Commit Graph

1200 Commits

Author SHA1 Message Date
Adrian Grange
075b11f004 Merge "Changed name of rd_pick_intra4x4mby_modes" 2013-08-02 11:36:46 -07:00
Dmitry Kovalev
741537f3ce Cleanup: replacing xd->seg with seg, and xd->lf with lf.
Change-Id: I73b59d7699a8e7e7acd3bf8041cb6c98ce9ba4bf
2013-08-01 15:38:16 -07:00
Dmitry Kovalev
ce8dedc353 Cleanup: removing unused function arguments.
Change-Id: I27471768980fc631916069f24bc7c482a5c9ca17
2013-08-01 13:41:38 -07:00
Dmitry Kovalev
b621e2d72e Nice looking motion vector clamping functions.
Removing assign_and_clamp_mv function, making implementation of clamp_mv
and clamp_mv2 more clear and consistent.

Change-Id: Iecd08e1c1bf0379f8314ebe01811f8253f4ade58
2013-08-01 13:40:26 -07:00
Adrian Grange
89e73c63c0 Fixed typos and added a few explanatory comments
Change-Id: Ib4e4b41094b54874ee34343dd77c0c131ceed9d2
2013-08-01 09:23:49 -07:00
Adrian Grange
5271d47892 Changed name of rd_pick_intra4x4mby_modes
The function name rd_pick_intra4x4mby_modes is confusing, so
I changed it to rd_pick_intra_sub_8x8_y_modes to better
reflect what the function does. Also added const qualifiers
to some of the input parameters and removed camel-case.

Change-Id: I23d53d4c7af5d79ed8a471acd59a09bbb47add39
2013-08-01 09:23:49 -07:00
Dmitry Kovalev
9239e96536 Removing get_mi_{row, col} functions.
Passing mi_row and mi_col parameters to functions explicitly. Removing
unused xd argument from scale_mv function.

Change-Id: Icb4c495ec72d26fb066c14470d3ae0b741fbf18a
2013-07-31 14:06:55 -07:00
Dmitry Kovalev
500ade243a Removing unused "ishp" arguments.
Using different variable names "allow_hp" and "use_hp" instead of "usehp".

Change-Id: I0cd5996ddeb46bd754473b680a993c0aaf8eb879
2013-07-31 11:27:53 -07:00
Adrian Grange
fbd73648dd Merge "Cleanup typos, remove unnecessary lines, replace switch" 2013-07-30 12:59:46 -07:00
Adrian Grange
b30a06b930 Cleanup typos, remove unnecessary lines, replace switch
Removed unnecessary code lines, replaced switch with an if,
fixed spelling errors and formatting.

Change-Id: Ie48aa4604aa0ed48362ca359d792fb21b2ec1dc6
2013-07-30 12:10:32 -07:00
Dmitry Kovalev
730a34416f Renaming NB_TXFM_MODES constant to TX_MODES.
Change-Id: I10bf06e3a3d5271221ae6a42a36074d01d493039
2013-07-29 13:38:40 -07:00
Dmitry Kovalev
23391ea835 Renaming TX_SIZE_MAX_SB to TX_SIZES.
Change-Id: I6aa4191935aa93461a07c41b59fdae1eb5f5f107
2013-07-29 12:25:34 -07:00
Ronald S. Bultje
118ccdcd30 Inverse dimension order in token_cost array.
This allows us to increment the position at the band-level only as
we go from one band to the next; more importantly, that allows us to
use an add instead of multiply instruction, and omit the instruction
altogether if the band doesn't change from one coef to the next, thus
being slightly faster (probably more noticeable on systems where a
multiply is expensive, like arm).

Change-Id: I4343fe35b9f9a47fa00b217bdcbf5f91ff96c381
2013-07-26 17:30:04 -07:00
Ronald S. Bultje
dcacce6dd9 Merge "Save pixels instead of coefficients in intra4x4 RD loop." 2013-07-26 17:20:58 -07:00
Ronald S. Bultje
d30c8f41ef Merge "Add best_rd breakout in intra4x4 RD loop." 2013-07-26 17:20:51 -07:00
Dmitry Kovalev
c09b81719f Merge "General cleanups." 2013-07-26 13:59:39 -07:00
Yunqing Wang
52256cdbca Modify static threshold calculation
Used 3 * standard_deviation in internal threshold calculation
instead of fit curve. This actually approached the algorithm
better.
For comparison, similar tests were done:
The overall psnr loss is less than before.
1. derf set:
when static-thresh = 1, psnr loss is 0.329%;
when static-thresh = 500, psnr loss is 0.970%;
2. stdhd set:
when static-thresh = 1, psnr loss is 0.922%;
when static-thresh = 500, psnr loss is 1.307%;

Similar speedup is achieved. For example,
clip            bitrate  static-thresh psnr    time
akiyo(cif)       500        0          48.952  5.077s(50f)
akiyo            500        500        48.866  4.169s(50f)

parkjoy(1080p)   4000       0          30.388  78.20s(30f)
parkjoy          4000       500        30.367  70.85s(30f)

sunflower(1080p) 4000       0          44.402  74.55s(30f)
sunflower        4000       500        44.414  68.69s(30f)

Change-Id: Ic78833642ce1911dbbd1cb6c899a2d7e2dfcc1f3
2013-07-25 19:59:33 -07:00
Yunqing Wang
845fd5011c Merge "Add encoding option --static-thresh" 2013-07-25 14:58:00 -07:00
Yunqing Wang
d36852b702 Add encoding option --static-thresh
This option exists in VP8, and it was rewritten in VP9 to support
skipping on different partition levels. After prediction is done,
we can check if the residuals in the partition block will be all
quantized to 0. If this is true, the skip flag is set, and only
prediction data are needed in reconstruction. Based on DCT's energy
conservation property, the skipping check can be estimated in
spatial domain.

The prediction error is calculated and compared to a threshold.
The threshold is determined by the dequant values, and also
adjusted by partition sizes. To be precise, the DC and AC parts
for Y, U, and V planes are checked to decide skipping or not.

Test showed that
1. derf set:
when static-thresh = 1, psnr loss is 0.666%;
when static-thresh = 500, psnr loss is 1.162%;
2. stdhd set:
when static-thresh = 1, psnr loss is 1.249%;
when static-thresh = 500, psnr loss is 1.668%;

For different clips, encoding speedup range is between several
percentage and 20+% when static-thresh <= 500. For example,
clip            bitrate  static-thresh psnr    time
akiyo(cif)       500        0          48.923  5.635s(50f)
akiyo            500        500        48.863  4.402s(50f)

parkjoy(1080p)   4000       0          30.380  77.54s(30f)
parkjoy          4000       500        30.384  69.59s(30f)

sunflower(1080p) 4000       0          44.461  85.2s(30f)
sunflower        4000       500        44.418  78.1s(30f)

Higher static-thresh values give larger speedup with larger
quality loss.

Change-Id: I857031ceb466ff314ab580ac5ec5d18542203c53
2013-07-25 14:28:05 -07:00
Dmitry Kovalev
7131cb0e3d General cleanups.
Removing unused constants, macros, and function declarations. Using
ROUND_POWER_OF_TWO macro, vp9_zero, vp9_copy where possible. Moving
#include from *.h to *.c. Merging for loops for motion vectors.

Change-Id: Ic3bf841764a2bb177128bb3a6d7aa8f68229cd13
2013-07-25 14:13:48 -07:00
Adrian Grange
e862c6f9eb Merge "Simplify handling of sub-partition motion vectors" 2013-07-25 12:58:38 -07:00
Adrian Grange
6f0f0e4907 Merge "Use local variables rather than structure members" 2013-07-25 12:57:52 -07:00
Adrian Grange
be700e140a Simplify handling of sub-partition motion vectors
Simplified the code that extracts and uses the motion
vectors for the 4 sub-partitions in rd_pick_partition.

Change-Id: Iaf698ef7ee3aef9edd59015e1ae065dd359b17d9
2013-07-25 11:51:51 -07:00
Dmitry Kovalev
fcc34796d2 Removing CONFIG_BALANCED_COEFTREE experiment.
Change-Id: I61a8b0101eac3ee2e0621d56151b90c269fd4db4
2013-07-24 15:53:42 -07:00
Dmitry Kovalev
9139ee0908 Adding condition inside get_tx_type_{4x4, 8x8, 16x16}.
Adding plane type check condition because it was always used outside of
get_tx_type_{4x4, 8x8, 16x16}.

Change-Id: I02f0bbfee8063474865bd903eb25b54d26e07230
2013-07-24 12:55:45 -07:00
Adrian Grange
4cfd36d8fd Use local variables rather than structure members
Although local copies of the mode member variables
(mode, ref_frame) were made, they were not used in
all places. Also, made a local copy of the
second_ref_frame member.

Change-Id: I84d8c822e5cb3d8a02fc3de8a4037ca3fea8bfad
2013-07-24 11:17:44 -07:00
Ronald S. Bultje
7817d3221f Save pixels instead of coefficients in intra4x4 RD loop.
Prevents doing duplicate IDCTs; encoding of first 50 frames of bus
(speed 0) @ 1500kbps goes from 1min4.0 to 1min3.5, i.e. 0.87% faster
overall.

Change-Id: I2df39e29ed9d5ea5e7d2704a34940ba622832ddd
2013-07-24 09:03:20 -07:00
Ronald S. Bultje
b72ecbb1b9 Add best_rd breakout in intra4x4 RD loop.
Encoding time of first 50 frames of bus (speed 0) @ 1500kbps goes from
1min5.4 to 1min4.0, i.e. 2.2% faster overall.

Change-Id: I8c32f2aff9a649ce7dd49d910dc5ba16b99c3bc6
2013-07-24 09:02:05 -07:00
Ronald S. Bultje
47336afd8d Merge "More optimizations for cost_coeffs()." 2013-07-23 21:36:12 -07:00
Dmitry Kovalev
db7f5d28b9 Removing vp9_is_interpolating_filter array.
All filters are interpolating now, so we don't need this array, all
values from this array are evaluated to true.

Change-Id: I9af6d8219ae0eb984063cd15e4e2296374ae4961
2013-07-23 14:24:39 -07:00
Dmitry Kovalev
2855d8aea1 Merge "Adding update_tx_counts function." 2013-07-23 13:57:59 -07:00
James Zern
8dede954c7 Merge "vp9: make some static tables const" 2013-07-23 11:37:01 -07:00
Jim Bankoski
86a9dec73c clean up bw, bh
many structures use bw and bh and they have different meanings.   This cl attempts
to start this clean up and remove unneccessary 2 step look up log and then
shift operations...

also removed partition type multiple operation code in bitstream.c.

Change-Id: I7e03e552bdfc0939738e430862e3073d30fdd5db
2013-07-23 06:51:44 -07:00
Paul Wilkins
7c134bc0cd Merge "Reworked the auto_mv_step_size speed feature" 2013-07-23 04:49:55 -07:00
James Zern
3c8cce353f vp9: make some static tables const
Change-Id: I8bcae51271673da8755c66a51aea005dfe6a3739
2013-07-22 19:19:13 -07:00
Ronald S. Bultje
e20fcd9585 More optimizations for cost_coeffs().
4x4:    163 ->  123 cycles (33% faster)
8x8:    491 ->  399 cycles (23% faster)
16x16: 1889 -> 1763 cycles (7% faster)
32x32: 8311 -> 8180 cycles (1.6% faster)

Overall encoding time of first 50 frames of bus (speed 0) @ 1500kbps
goes from 1min4.33 to 1min3.00, i.e. 2.11% faster.

Change-Id: Ib52d1dbb5649b14de769d3e7a74af67440b5284f
2013-07-22 16:09:09 -07:00
Dmitry Kovalev
b2fc6fa969 Adding update_tx_counts function.
Moving common encoder/decoder code to update_tx_counts. Also renaming
vp9_get_pred_probs_tx_size to get_tx_probs2 and adding get_tx_probs to
call vp9_get_pred_context_tx_size inside read_selected_tx_size only once
(twice before).

Change-Id: Ia50247f3893de88ef8e9041b0d44be44a40aaa4d
2013-07-22 14:57:43 -07:00
Yaowu Xu
fc186dcad6 fix a build error
Change-Id: I3b05687f439ff6a7c426d2c97a6c58c831fa51ac
2013-07-22 12:37:30 -07:00
Jingning Han
416f315e82 Merge "Skip buffer update in sub8x8 rd loop" 2013-07-22 12:08:22 -07:00
Jingning Han
a5a9f5f7f3 Merge "Optimize operation flow in sub8x8 rd loop" 2013-07-22 12:08:15 -07:00
Jingning Han
409e77f2d4 Optimize operation flow in sub8x8 rd loop
Stack the rate-distortion statistics in the sub8x8 rd loop. This allows
the encoder to skip the forward transform, quantization, and coeff cost
estimation, in the sub8x8 rd optimization search, if the motion
vector(s) are of integer pixel value, and have been tested in the
previous prediction filter type rd loops of the same block.

This gives about 2% speed-up for bus_cif at 2000 kpbs, for speed 0.
Its efficacy depends how frequently the motion search will select an
integer motion vector.

Change-Id: Iee15d4283ad4adea05522c1d40b198b127e6dd97
2013-07-22 10:40:33 -07:00
Paul Wilkins
1d189d6464 Re-order mode search in rd.
Mode search order in rd loop changed to better reflect
observed hit counts.

Also some adjustment of the baseline mode rd thresholds
to reflect the order change and observed frequencies.

Change-Id: I47a131cc83e11551df8add6d6d8d413d78d3a63c
2013-07-22 17:21:12 +01:00
Dmitry Kovalev
39342db138 Merge "Consistent names for inter mode probabilities and encodings." 2013-07-20 22:40:51 -07:00
Dmitry Kovalev
f66821afbb Merge "Removing frame_type field from MACROBLOCKD struct." 2013-07-20 22:40:06 -07:00
Jingning Han
c725502bf3 Skip buffer update in sub8x8 rd loop
This commit allows the encoder to skip a few buffer update steps in
rd_pick_best_mbsegmentation, when early breakout has been triggered
in the rd_check_segment_txsize. It provides about 1% speed-up for
bus_cif at 2000 kbps, in the settings of speed 0.

Change-Id: Ica034f10a24dec572b397d8389a2b81020ebc0b9
2013-07-20 21:38:12 -07:00
Deb Mukherjee
302698fb12 Reworked the auto_mv_step_size speed feature
This patch modifies the auto_mv_step_size speed feature to
use a combination of the maximum magnitude mv from the last
inter frame, and the maximum magnitude mv for the two reference
mvs with the same reference. For arf frames, the max mav step
for the resolution is used.
The bounds therefore are slightly tighter. The feature is made
a speed 1 feature.

Rebased.

Results (when this feature is turned on over speed 0):
derfraw300: -0.046% psnr, about 5+% speedup
(tested on football: goes from 4m30.760s to 4m17.410s).

Change-Id: If492797a61b0b4b3e58c0b8f86afb880165fc9f6
2013-07-19 15:12:56 -07:00
Dmitry Kovalev
e71a4a77bb Merge "Renaming TXFM_MODE to TX_MODE (like TX_SIZE, TX_TYPE)." 2013-07-19 12:14:32 -07:00
Dmitry Kovalev
97e96bc4e9 Removing frame_type field from MACROBLOCKD struct.
Change-Id: Ia4e83913251c1cdc7aa2abd64bf01ecb1a962119
2013-07-19 11:55:36 -07:00
Dmitry Kovalev
c0eb57406c Renaming TXFM_MODE to TX_MODE (like TX_SIZE, TX_TYPE).
Moving TX_MODE enum to vp9_enums.h. Renaming txfm_mode variables to
tx_mode.

Change-Id: I459d1af6dd928ce7fccdf8ce30b6f1ca057bef92
2013-07-19 11:37:13 -07:00
Dmitry Kovalev
afe43d4089 Removing redundant VP9_COMMON* from function signatures.
Functions: vp9_get_pred_context_switchable_interp,
           vp9_get_pred_context_intra_inter,
           vp9_get_pred_context_single_ref_p1,
           vp9_get_pred_context_single_ref_p2.

Change-Id: I3d6fb8aee23c9062270768e1e6da416dd9bb8f96
2013-07-19 11:20:49 -07:00
Dmitry Kovalev
bc7acb134b Consistent names for inter mode probabilities and encodings.
Renaming vp9_sb_mv_ref_tree to vp9_inter_mode_tree, and
vp9_sb_mv_ref_encoding_array to vp9_inter_mode_encodings.

Change-Id: I0e91fbf81350d3ec5a2599064c74089b5d06133a
2013-07-19 10:40:04 -07:00
Yaowu Xu
37d901a47a Merge "Add best_rd breakout to keyframe partition selection also." 2013-07-18 17:50:39 -07:00
Yaowu Xu
67fb0679ee Merge "Merge scale_factors and scale_factors_uv." 2013-07-18 17:50:34 -07:00
Yaowu Xu
55b52e32da Merge "Do in-place UV intra mode selection." 2013-07-18 17:50:07 -07:00
Yaowu Xu
51972d1279 Merge "Change break statement in a 2d loop to a return statement." 2013-07-18 17:49:58 -07:00
Dmitry Kovalev
92f4198d52 Merge "Using VP9_REF_NO_SCALE instead of (1 << VP9_REF_SCALE_SHIFT)." 2013-07-18 17:29:05 -07:00
Dmitry Kovalev
0b562b2d3d Using VP9_REF_NO_SCALE instead of (1 << VP9_REF_SCALE_SHIFT).
Change-Id: Ide58a74d31ff948319445a6337d2c05e98720e34
2013-07-18 15:12:46 -07:00
Ronald S. Bultje
96e4db2660 Add best_rd breakout to keyframe partition selection also.
Change-Id: I96b8058f6dfecf8aa3e152cdcbfd7e10071fbbc9
2013-07-18 14:10:56 -07:00
Ronald S. Bultje
5ebe503f04 Merge scale_factors and scale_factors_uv.
This prevents a duplicate memcpy of a 128-byte struct every time
set_scale_factors() is called (which is a lot), thus leading to a
decrease from 3.7 MB to 1.85 MB of struct copying per 64x64 block
RD/partition loop.

Overall, this decreases encoding time of the first 50 frames of bus
@ 1500kbps (speed 0) from 1min5.9 to 1min4.9, i.e. about a 1.5%
overall speedup. We can likely get more gains by removing the copy
of the other struct (and replacing it with an indexing) as well.

Change-Id: I3dceb7e79f71e6fe911b11cc994cf89a869dde7a
2013-07-18 14:10:56 -07:00
Ronald S. Bultje
df4b4fab26 Do in-place UV intra mode selection.
This means we only do UV intra mode selection if we find any intra
mode to actually be useful at all; in addition, we only do UV intra
mode selection for the transform sizes that were selected, rather
than all sizes available in this partition.

First 50 frames of bus @ 1500kbps (speed 0) gains about 5% with this
change.

Change-Id: I7b461eb8b803247f57896c5a9505f745b55502b3
2013-07-18 14:10:56 -07:00
Ronald S. Bultje
e54a5782b9 Change break statement in a 2d loop to a return statement.
The break statement only breaks out of the nested loop, not the
top-level loop, so it doesn't always work as intended. Changing it
to a return statement does what's intended.

Change-Id: I585419823b39a04ec8826b1c8a216099b1728ba7
2013-07-18 14:10:56 -07:00
Ronald S. Bultje
2d4929e340 Remove motion vectors from PARTITION_INFO.
The same information already exists in union b_mode_info.

Change-Id: Iac5086b99a3c3cc270380138062bb693e58f9e6d
2013-07-18 14:10:52 -07:00
Ronald S. Bultje
9da67da04a Merge "Fix bug where we don't choose any mode in RD selection." 2013-07-18 12:47:50 -07:00
Ronald S. Bultje
247197d57b Fix bug where we don't choose any mode in RD selection.
This could happen during golden overlay frame coding from a previous
alt-ref frame if the special overlay code was triggered.

Change-Id: I3056d0c547cd26903b260ef93c94026e96bd9868
2013-07-18 12:13:15 -07:00
Ronald S. Bultje
4f5815290c Merge "Fix bug which skips zeromv even if near/nearest is not 0,0." 2013-07-18 10:06:51 -07:00
Ronald S. Bultje
deb7456058 Fix bug which skips zeromv even if near/nearest is not 0,0.
Change-Id: Id4f454831f3f11099f39c30246adeaa52857d08d
2013-07-18 09:35:19 -07:00
Jingning Han
ced3c20165 Use mv_check_bounds in sub8x8 rd loop
Make the use of mv_check_bounds consistent for mvs of both ref_frame[0]
and ref_frame[1].

Change-Id: I1ca24865cc7232ca9cbe5db566c53abad1592211
2013-07-17 17:13:51 -07:00
Ronald S. Bultje
facecd80da Merge "Add a best_yrd shortcut in splitmv mode search." 2013-07-17 16:11:13 -07:00
Ronald S. Bultje
056111c822 Merge "Skip redundant nearest/near/zero encodes in splitmv." 2013-07-17 16:10:51 -07:00
Ronald S. Bultje
0b1eba25b2 Merge "Skip nearest/near/zero redundant encodes." 2013-07-17 16:10:41 -07:00
Ronald S. Bultje
607424449c Merge "Best_rd breakout in rd partition search." 2013-07-17 16:10:22 -07:00
Yaowu Xu
6ac5b7db2c Merge "changed mode checking order" 2013-07-17 14:44:40 -07:00
Dmitry Kovalev
a7a1e96136 Merge changes Ieffea49e,Idf610746
* changes:
  Removing two unused arguments from vp9_inc_mv signature.
  Changing signature of vp9_get_pred_probs_tx_size.
2013-07-17 14:44:20 -07:00
Ronald S. Bultje
c6917528a5 Add a best_yrd shortcut in splitmv mode search.
Encoding of first 50 frames of bus (speed 0) @ 1500kbps goes from
1min6.2 to 1min5.9, i.e. 0.5% faster overall.

Change-Id: I59d8a3b2f0a75010fa041d5e2646c8caac5bd683
2013-07-17 14:21:57 -07:00
Ronald S. Bultje
161c995658 Skip redundant nearest/near/zero encodes in splitmv.
Encode of first 50 frames of bus @ 1500kbps (speed 0) goes from
1min7.3 to 1min6.2, i.e. 1.7% faster overall.

Change-Id: I19d2deacfbffadd61d32551cee9586757ab4a987
2013-07-17 13:53:48 -07:00
Yaowu Xu
42facc292d changed mode checking order
Change-Id: Ic4c4b363ed840935e42f495f13ea5e601a56f1b2
2013-07-17 13:43:50 -07:00
Ronald S. Bultje
8fea880b6f Skip nearest/near/zero redundant encodes.
Encode of first 50 frames of bus @ 1500kbps (speed 0) goes from 1min12.8
to 1min7.3, i.e. 8% faster.

Change-Id: Ia22d1c7b687316c553cc60eacae988b24e175b62
2013-07-17 11:33:15 -07:00
Ronald S. Bultje
9f427bfe98 Best_rd breakout in rd partition search.
About 15% faster for bus (speed 0) first 50 frames @ 1500kbps, which
goes from 1min36 to 1min24. Results become slightly better (+0.2% on
derf/yt, +0.4% on hd), probably because of a bugfix for skipmode in
super_block_yrd(). Overall speed change (on derfraw300) is roughly
-13%. This can probably be improved further by caching best_yrd
between partition searches. Also, we might be able to get more
speedups by always doing PARTITION_NONE before PARTITIONS_SPLIT, not
just at the sb8x8 level.

Change-Id: I83736949ebd5b4a3b400ee688d7661913fefc98b
2013-07-17 09:56:46 -07:00
Ronald S. Bultje
83c7e13a6b Do a skip-block check for sub8x8 partitions also.
+0.2% SSIM and glbPSNR on derfraw300.

Change-Id: I9cba0bca55e606a22f557c7732b064f738efe84d
2013-07-17 09:46:47 -07:00
Yunqing Wang
df90d58f4f Speed up motion estimation using small partitions' result(experiment)
Current partition checking starts from small sizes, and then goes up
to large sizes. This experiment uses the small partitions' motion
estimation result, which is already available, to speed up the
large partition's motion estimation. We can decide to skip some
patition checkings if they are unlikely choices. We could use the
motion vector(MV) result as current partition's prediction MV, limit
the search range and reference frame.

Current result at speed 1:
psnr loss: 1.19% for stdhd, 0.287% for derf.
speed gain: 14% for sunflower(hd), 11% for akiyo.

Further improvement will be done later.

Change-Id: I5abfd070e9cace2e91e2a0247d1325df313887ab
2013-07-17 09:11:47 -07:00
Paul Wilkins
d66eab15dd Merge "Move uv intra mode selection in rd loop." 2013-07-17 05:19:26 -07:00
Paul Wilkins
154c34a3ee Merge "Limit transform sizes searched for uv intra." 2013-07-17 03:40:11 -07:00
Paul Wilkins
2ee338ce3b Move uv intra mode selection in rd loop.
Use an estimate based on DC_PRED for intra uv cost
within the rd loop then only do a full uv mode analysis
if an intra mode is chosen.

Significant speed gains in some cases. Currently only
enabled for speed 2 pending speed/quality tests.

Change-Id: Ie851a12400d5483bce47ec0e3ccb8516041e91c0
2013-07-17 11:11:21 +01:00
Paul Wilkins
6c667f0ffe Limit transform sizes searched for uv intra.
Apply limit if search_method == USE_LARGESTALL
to the range of UV tx sizes searched.

Change-Id: I6db29f0dd237285ffc50d75a37e8b68151ad821c
2013-07-17 11:08:55 +01:00
Paul Wilkins
5f4722c75f Merge "Minor cleanup in code to fine uv tx_size." 2013-07-17 02:50:09 -07:00
Jingning Han
a142d6fc93 Skip redundant motion search in 4x4 level rd loop
This commit makes the encoder to perform motion search only once
per reference frame type for each 4x4/4x8/8x4 block. For bus_cif
at 2000 kbps, the runtime goes from 253812ms -> 217817ms
(14% speed-up) for speed 0.

Change-Id: I5f17599ccc8cfaf93ccb4f98fcb6008af6d79e92
2013-07-16 17:21:11 -07:00
Dmitry Kovalev
5b65a71cdc Changing signature of vp9_get_pred_probs_tx_size.
Removing VP9_COMMON* argument and adding struct tx_probs* instead of
MACROBLOCKD*.

Change-Id: Idf61074631a90ec51eac22c8dcd977f44ac0757c
2013-07-16 16:34:54 -07:00
Paul Wilkins
30d2ea45ce Minor cleanup in code to fine uv tx_size.
Change-Id: I94b97a966b5efbc9a243048f1f5ddbbdc4b1846e
2013-07-16 18:27:33 +01:00
Dmitry Kovalev
ca75f1255f Removing and moving around constant definitions.
Removing unused and duplicated constants, moving them from *.h to *.c
if possible.

Change-Id: Ief4d6b984a3ca2e9b38504f0d855ed072cf7133f
2013-07-15 19:26:30 -07:00
Jingning Han
faff6ed0fb Skip duplicate block encoding in the rd loop
This speed feature allows the encoder to largely remove the spatial
dependency between blocks inside a 64x64 superblock, thereby removing
the need to repeatedly encode superblocks per partition type in the
rate-distortion optimization loop.

A major challenge lies in the intra modes tested in the rate-distortion
optimization loop. The subsequent blocks do not have access to the
reconstructed boundary pixels without the intermediate coding steps.
This was resolved by using the original pixels for intra prediction
in the rd loop, followed by an appropriately designed distortion
modeling on the quantization parameters. Experiments also suggested
that the performance impact is more discernible at lower bit-rate/psnr
settings. Hence a quantizer dependent threshold is applied to deactivate
skip of block coding.

For bus_cif at 2000 kbps,
speed 0: runtime 269854ms -> 237774ms (12% speed-up) at 0.05dB
         performance loss.

speed 1: runtime 65312ms  -> 61536ms, (7% speed-up) at 0.04dB
         performance loss.

This operation is currently turned on in settings of speed 1.

Change-Id: Ib689741dfff8dd38365d8c1b92860a3e176f56ec
2013-07-15 11:08:58 -07:00
Yaowu Xu
fb754b182f Fix a build issue
Change-Id: I23a75c495ed7ea917d7f312bef0990e20a6b53d9
2013-07-12 11:38:44 -07:00
Deb Mukherjee
94c481f9f1 Some minor cleanups for efficiency
Implements some of the helper functions more efficiently with
lookups rathers than branches. Modeling function is consolidated
to reduce some computations.

Also merged the two enums BLOCK_SIZE_TYPES and BlockSize into
one because there is no need to keep them separate (even though
the semantics are a little different).

No bitstream or output change.

About 0.5% speedup

Change-Id: I7d71a66e8031ddb340744dc493f22976052b8f9f
2013-07-12 10:22:56 -07:00
Ronald S. Bultje
ee09dd9949 Remove unused function block_error().
Change-Id: I78a79fc51c2d7cc3c261f35b569155397f3dc0c4
2013-07-11 17:14:03 -07:00
Dmitry Kovalev
8c05e59065 Calling is_inter_mode() instead of custom code.
Change-Id: Iccd4ab95ea51a6d57ed43947f2fd7ad92e8979cf
2013-07-11 14:14:47 -07:00
Dmitry Kovalev
c4ad3273c7 Moving segmentation related vars into separate struct.
Adding segmentation struct to vp9_seg_common.h. Struct members are from
macroblockd and VP9Common structs. Moving segmentation related constants
and enums to vp9_seg_common.h.

Change-Id: I23fabc33f11a359249f5f80d161daf569d02ec03
2013-07-11 11:57:57 -07:00
Jingning Han
18803f9cc4 Fix tx_type bug in intra4x4 rd loop
This commit fixed the mis-use of the tx_type for inverse transform
in intra4x4 rate-distortion optimization loop. It improves the
overall coding performance.

Change-Id: I7fe9953175b74890357dbcee33c138573766e980
2013-07-10 15:49:49 -07:00
Deb Mukherjee
7494bba66b Merge "Prunes out full-rd computation based on modeled rd" 2013-07-10 15:37:11 -07:00
Jim Bankoski
865ca76604 Merge "remove warnings when NDEBUG is set" 2013-07-10 14:39:39 -07:00
Jim Bankoski
6591cf2f7e remove warnings when NDEBUG is set
Change-Id: Ie0cb732fdcb98616a422c4463bff80642248d136
2013-07-10 14:27:20 -07:00
Deb Mukherjee
53ff43adc3 Prunes out full-rd computation based on modeled rd
Adds a speed feature to eliminate full-rd computation if the modeled
rd or rd based on a different parameter in the same mode is already
a lot larger than the best rd yet.

Specifically, only search the sharp and smooth filters if the modeled
rd cost based on the  regular filter is within a certain factor of the
best rd cost so far. Also, skip full-rd computation of non splitmv
inter modes if the modeled rd cost based on pred error is within the
same factor of the best rd cost so far.

Also adds some enhancements in the rd search for splitmv mode to
speed things up by early breakouts. Negligible impact on performance.

Resuts on derfraw300:
psnr:    -0.013% with the splitmv enhancements, -0.24% with the rd
         breakout feature on.
speedup: 6% with splitmv enhancements, 20% with also residual breakout
         (tested on football sequence at 600 Kbps)

Change-Id: I37abc308ea9f110c1679ce649b6a7e73ab1ad5fc
2013-07-10 13:49:49 -07:00
Yaowu Xu
e52eec490c Merge "Add a feature to reduce chrome intra mode search" 2013-07-10 11:35:47 -07:00
Ronald S. Bultje
b1df674a99 Remove memcpy() in handle_inter_mode() filter selection.
Encode time of first 50 frames of bus (speed 0) @ 1500kbps goes from
2min4.9 to 2min3.1, i.e. a 1.4% speedup overall.

Change-Id: I9b25e87974430cb942caa276410bb2eda815bd83
2013-07-10 09:27:56 -07:00
Yaowu Xu
bed27a960a Add a feature to reduce chrome intra mode search
Change-Id: I721ebdeef2b53ce3e5c3eba3f7462ae2103c95a8
2013-07-10 08:59:18 -07:00
Jim Bankoski
fb027a7658 removing case statements around prediction entropy coding
Removes SEG_ID
Removes MBSKIP
Removes SWITCHABLE_INTERP
Removes INTRA_INTER
Removes COMP_INTER_INTER
Removes COMP_REF_P
Removes SINGLE_REF_P1
Removes SINGLE_REF_P2
Removes TX_SIZE

Change-Id: Ie4520ae1f65c8cac312432c0616cc80dea5bf34b
2013-07-09 20:10:16 -07:00
Yaowu Xu
205efbc153 Revert "Remove memcpy() in handle_inter_mode() filter selection."
This reverts commit fcf7998a47.

Change-Id: Ic6532223faec9f1483b78adb2e37b79c7b1a0efb
2013-07-09 17:42:10 -07:00
Ronald S. Bultje
204d1b7058 Merge "Unbreak lossless." 2013-07-09 09:54:48 -07:00
Ronald S. Bultje
059c0ba5d4 Unbreak lossless.
Change-Id: I8130ec9b5371c65e885f245a5ac73840c23cb4a1
2013-07-09 09:46:37 -07:00
Dmitry Kovalev
1c65c580d6 Merge "Refactoring setup_pre_planes function." 2013-07-08 20:08:05 -07:00
Ronald S. Bultje
8fde07a3ae Don't recalculate mv_ref costs for each block/partition.
Changes cost_mv_ref() into doing a LUT into pre-calculated cost
arrays instead. Encode time of first 50 frames of bus (speed 0)
@ 1500kbps goes from 2min11.6 to 2min10.9, i.e. 0.5% faster overall.

Change-Id: If186e92c34c201b29cbbc058785a15c9c09e433a
2013-07-08 16:22:39 -07:00
Ronald S. Bultje
fcf7998a47 Remove memcpy() in handle_inter_mode() filter selection.
Encode time of first 50 frames of bus (speed 0) @ 1500kbps goes from
2min4.9 to 2min3.1, i.e. a 1.4% speedup overall.

Change-Id: Ibe8b08d159797504c5d0c5122de1b6da3b6595e0
2013-07-08 16:22:39 -07:00
Ronald S. Bultje
ed995afba1 Make frame-wide filter-type decision fully RD-based.
Overall, on all test sets, this gains about +0.2% on all metrics.
City is a clip where this really hurts (-1.0% on all metrics), I'm
not quite sure why yet. Maybe interesting to look into in the future.

Change-Id: I6f0eecb20e72f0194633270d30bf00d76d9eae78
2013-07-08 16:22:37 -07:00
Deb Mukherjee
d9b62160a0 Implements several heuristics to prune mode search
Skips mode searches for intra and compound inter modes depending
on the best mode so far and the reference frames. The various
heuristics to be used are selected by bits from a flag. The
previous direction based intra mode search pruning is also absorbed
in this framework.

Specifically the flags and their impact are:

1) FLAG_SKIP_INTRA_BESTINTER (skip intra mode search for oblique
directional modes and TM_PRED if the best so far is
an inter mode)
derfraw300: -0.15%, 10% speedup

2) FLAG_SKIP_INTRA_DIRMISMATCH (skip D27, D63, D117 and D153
mode search if the best so far is not one of the closest
hor/vert/diagonal directions.
derfraw300: -0.05%, about 9% speedup

3) FLAG_SKIP_COMP_BESTINTRA (skip compound prediction mode
search if the best so far is an intra mode)
derfraw300: -0.06%, about 7-8% speedup

4) FLAG_SKIP_COMP_REFMISMATCH (skip compound prediction search
if the best single ref inter mode does not have the same ref
as one of the two references being tested in the compound mode)
derfraw300: -0.56%, about 10% speedup

Change-Id: I1a736cd29b36325489e7af9f32698d6394b2c495
2013-07-08 12:17:12 -07:00
Paul Wilkins
ef0ca2deaa Merge "Fix to comp_inter_joint_search_thresh feature." 2013-07-04 03:27:00 -07:00
Dmitry Kovalev
f72e072555 Refactoring setup_pre_planes function.
Removing set_refs, adding set_ref function.

Change-Id: I5635c478b106ae4e57d317f1c83d929644307e63
2013-07-03 17:42:01 -07:00
Jingning Han
68172dbede Merge "Enable early termination in rd search" 2013-07-03 14:20:41 -07:00
Jingning Han
2bd6fe08f8 Enable early termination in rd search
This commit allows encoder to detect the cumulative rate-distortion
cost per transformed block inside a partition. If the cumulative
rd cost is already above the best rd value, it terminates the rest
operations and continue to next prediction mode test.

It reduces the runtime of bus at target bit-rate 2000 from 308 second
to 266 second, i.e., about 13% speed-up at no performance penalty.

Change-Id: I5f15a3d8955d97031d5653006027866a00654e7a
2013-07-03 12:54:18 -07:00
Paul Wilkins
f58b44ad62 Fix to comp_inter_joint_search_thresh feature.
When this is 0 (BLOCK_SIZE_AB4X4) we want to do
the inter joint search for all sizes.

Change-Id: Id40cd6fe7790e7e1165352b9cef5e12fa8c0bc88
2013-07-03 16:58:34 +01:00
Paul Wilkins
72c5778ec5 Added two new skip experiments.
sf->unused_mode_skip_lvl. Tests modes as normal for all
sizes at or below the given level. At larger sizes it skips
all modes that were not chosen at any smaller size.
Hence setting BLOCK_SIZE_SB64X64 is in effect off.
Setting BLOCK_SIZE_AB4X4 will only consider modes that
were chosen for one or more 4x4 blocks at larger sizes.

sf->reference_masking.
Do a test encode of the NONE partition at one size and create
a reference frame mask based on the best rd choice. In the
full search only allow this reference frame.
Currently it is testing 64x64 and repeats this in the full search.
This does not work well with Jim's Partition code just now and
is disabled by default.

Change-Id: I8f8c52d2ef4a0c08100150b0ea4155d1aaab93dd
2013-07-03 16:56:06 +01:00
Dmitry Kovalev
be77f6bbbf Removing redundant struct from union b_mode_info.
Change-Id: I08fc6e474ff2c12cfa065bae4989c724276e2c83
2013-07-02 16:51:57 -07:00
Deb Mukherjee
37501d687c Speed feature to binary search dir intramodes
This speed feature will skip searching the directional intra prediction
modes D63, D117, D27, D153 if the best intra mode so far is not one of
the diagonal, horizontal or vertical directions closest to the respective
directions being tested. In other words, this implements a sort of
binary search in the angular domain.

Speedup: about 9-10%
Results: -0.05% only on derfraw300.

Change-Id: I413584c41f2a3e8dabfbdeb40718c8fc4b1d63a2
2013-07-02 14:07:19 -07:00
Deb Mukherjee
8d3d2b76f3 Tx size selection enhancements
(1) Refines the modeling function and uses that to add some speed
features. Specifically, intead of using a flag use_largest_txfm as
a speed feature, an enum tx_size_search_method is used, of which
two of the types are USE_FULL_RD and USE_LARGESTALL. Two other
new types are added:
USE_LARGESTINTRA (use largest only for intra)
USE_LARGESTINTRA_MODELINTER (use largest for intra, and model for
inter)

(2) Another change is that the framework for deciding transform type
is simplified to use a heuristic count based method rather than
an rd based method using txfm_cache. In practice the new method
is found to work just as well - with derf only -0.01 down.
The new method is more compatible with the new framework where
certain rd costs are based on full rd and certain others are
based on modeled rd or are not computed. In this patch the existing
rd based method is still kept for use in the USE_FULL_RD mode.
In the other modes, the count based method is used.
However the recommendation is to remove it eventually since the
benefit is limited, and will remove a lot of complications in
the code

(3) Finally a bug is fixed with the existing use_largest_txfm speed feature
that causes mismatches when the lossless mode and 4x4 WH transform is
forced.

Results on derf:
USE_FULL_RD: +0.03% (due to change in the tables), 0% encode time reduction
USE_LARGESTINTRA: -0.21%, 15% encode time reduction (this one is a
pretty good compromise)
USE_LARGESTINTRA_MODELINTER: -0.98%, 22% encode time reduction
(currently the benefit of modeling is limited for txfm size selection,
but keeping this enum as a placeholder) .
USE_LARGESTALL: -1.05%, 27% encode-time reduction (same as existing
use_largest_txfm speed feature).

Change-Id: I4d60a5f9ce78fbc90cddf2f97ed91d8bc0d4f936
2013-07-02 13:54:00 -07:00
Ronald S. Bultje
3cc6eb7c00 Merge "Make get_coef_context() branchless." 2013-07-02 11:48:15 -07:00
Jingning Han
b91a1586a3 Calculate rd cost per transformed block
Compute the rate-distortion cost per transformed block, and cumulate
the cost through all blocks inside a partition. This allows encoder
to detect if the cumulative rd cost is already above the best rd cost,
thereby enabling early termination in the rate-distortion optimization
search.

Change-Id: I0a856367a9a7b6dd0b466e7b767f54d5018d09ac
2013-07-02 09:58:46 -07:00
Paul Wilkins
b7cd01ed73 Revert "New motion threshold factor - speed feature."
This reverts commit 1377278180.
Also fixes a spelling mistake.

Change-Id: I5be8aa4d8d3c0323d4a6f41968a7b2c048949c3f
2013-07-02 15:06:40 +01:00
Ronald S. Bultje
26b6318de8 Make get_coef_context() branchless.
This should significantly speedup cost_coeffs(). Basically what the
patch does is to make the neighbour arrays padded by one item to
prevent an eob check in get_coef_context(), then it populates each
col/row scan and left/top edge coefficient with two times the same
neighbour - this prevents a single/double context branch in
get_coef_context(). Lastly, it populates neighbour arrays in pixel
order (rather than scan order), so we don't have to dereference the
scantable to get the correct neighbours.

Total encoding time of first 50 frames of bus (speed 0) at 1500kbps
goes from 2min10.1 to 2min5.3, i.e. a 2.6% overall speed increase.

Change-Id: I42bcd2210fd7bec03767ef0e2945a665b851df56
2013-07-01 16:34:10 -07:00
Yaowu Xu
ba3b2604f0 Merge "Quantize (64-bit only, for now) SSSE3 SIMD." 2013-07-01 15:58:57 -07:00
Ronald S. Bultje
7353ceab9d Quantize (64-bit only, for now) SSSE3 SIMD.
Total encoding time for first 50 frames of bus (speed 0) @ 1500kbps
goes 2min34.8 to 2min14.4, i.e. a 10.4% overall speedup. The code is
x86-64 only, it needs some minor modifications to be 32bit compatible,
because it uses 15 xmm registers, whereas 32bit only has 8.

Change-Id: I2df53770c2e850813ffa713e1a91b45b0082b904
2013-07-01 11:36:07 -07:00
Paul Wilkins
1377278180 New motion threshold factor - speed feature.
Added a speed feature that focuses only on thresholds
for new motion modes.

Moved sf->comp_inter_joint_search_thresh into speed
1.  This has ~+0.4% impact on quality at speed 0 as
our quality reference baseline.

Slight adjustment to baseline thresholds.

Change-Id: I7ebf104f1fe29af77ed4837b2e84be065621bbe5
2013-07-01 12:11:21 +01:00
Ronald S. Bultje
bc70c60b25 Merge "fixed a bug where sse is not populated" 2013-06-29 07:42:41 -07:00
Yaowu Xu
f853e662b7 fixed a bug where sse is not populated
Change-Id: I692d800af1f976c84a76f8bd66864c4b39540abc
2013-06-28 17:10:22 -07:00
Ronald S. Bultje
d00b8e5f82 Inline vp9_get_coef_context() (and remove vp9_ prefix).
Makes cost_coeffs() a lot faster:
4x4: 236 -> 181 cycles
8x8: 888 -> 588 cycles
16x16: 3550 -> 2483 cycles
32x32: 17392 -> 12010 cycles

Total encode time of first 50 frames of bus (speed 0) @ 1500kbps goes
from 2min51.6 to 2min43.9, i.e. 4.7% overall speedup.

Change-Id: I16b8d595946393c8dc661599550b3f37f5718896
2013-06-28 10:40:21 -07:00
Ronald S. Bultje
e3ce2b2ab3 Minor change to prevent one level of dereference in cost_coeffs().
4x4: 234 -> 236 cycles
8x8: 878 -> 888 cycles
16x16: 3664 -> 3550 cycles
32x32: 18134 -> 17392 cycles

Change-Id: I37a51bfbb0060a3a54f09c6045c14a989811ed78
2013-06-28 10:29:07 -07:00
Ronald S. Bultje
91d223bd5c Some minor optimizations for cost_coeffs().
Cycle timings for first 3 frames of bus (speed 0) at 1500kbps:
4x4: 298 -> 234 cycles
8x8: 1227 -> 878 cycles
16x16: 23426 -> 18134 cycles
32x32: 4906 -> 3664 cycles

Total encode time of first 50 frames of bus @ 1500kbps (speed 0) goes
from 3min0.7 to 2min51.6 seconds, i.e. 5.3% faster.

Change-Id: I68a0e1b530b0563b84a67342cca4b45146077e95
2013-06-28 10:29:02 -07:00
Ronald S. Bultje
af660715c0 Make coefficient skip condition an explicit RD choice.
This commit replaces zrun_zbin_boost, a method of biasing non-zero
coefficients following runs of zero-coefficients to be rounded towards
zero, with an explicit skip-block choice in the RD loop.

The logic is basically that if individual coefficients should be rounded
towards zero (from a RD point of view), the trellis/optimize loop should
take care of it. If whole blocks should be zero (from a RD point of
view), a single RD check is much more efficient than a complete
serialization of the quantization loop.

Quality change: derf +0.5% psnr, +1.6% ssim; yt +0.6% psnr, +1.1% ssim.
SIMD for quantize will follow in a separate patch. Results for other
test sets pending.

Change-Id: Ife5fa641163ac5150ac428011e87188f1937c1f4
2013-06-28 10:28:49 -07:00
Yaowu Xu
8b9eea0a34 Minor cleanups
Change-Id: I379617c1c731a686b3f7e032b8805860c1055b12
2013-06-28 09:19:50 -07:00
Paul Wilkins
05ffdf2625 Merge "Auto adapt step size feature." 2013-06-27 02:28:41 -07:00
Paul Wilkins
59af9049d3 Merge "Start adaptive threshold for each mode at max." 2013-06-27 02:28:36 -07:00
Paul Wilkins
5bcf069c6b Merge "Change meaning of cpi->sf.first_step and rename." 2013-06-27 02:28:21 -07:00
Jingning Han
fc1cfd8e32 Merge "Make intra predictor reference buffer configurable" 2013-06-26 19:02:02 -07:00
Jingning Han
861cb06c67 Make intra predictor reference buffer configurable
This commit enables configurable reference buffer pointer for intra
predictor. This allows later removal of spatial dependency between
blocks inside a 64x64 superblock in the rate-distortion optimization
loop.

Change-Id: I02418c2077efe19adc86e046a6b49364a980f5b1
2013-06-26 17:17:21 -07:00
Paul Wilkins
9f3ab83486 Auto adapt step size feature.
Also tweaks to other features and experiments with
what is on and off at different speed settings.

Change-Id: I3e1d0be0d195216bf17c2ac5df67f34ce0b306b2
2013-06-26 19:48:39 +01:00
Dmitry Kovalev
49dee16879 Merge "Using get_plane_block_{width, height} instead of custom code." 2013-06-26 10:23:27 -07:00
Paul Wilkins
689957e3ad Start adaptive threshold for each mode at max.
Each frame we reset all adaptive thresholds to MAX
rather than base. As modes are picked their thresholds
drop down.

Change-Id: Ia37f03a73003c2d9bfcda57edea07205e9a0e5e8
2013-06-26 17:04:47 +01:00
Paul Wilkins
e606cac046 Change meaning of cpi->sf.first_step and rename.
Renamed cpi->sf.first_step to cpi->sf.reduce_first_step_size
and changed its meaning such that it is a delta applied to
reduce the default first step size (>> x) in the motion search
rather than an absolute value.

The default first step size is already changed according to the image
dimensions (smaller for smaller images). cpi->sf.reduce_first_step_size
now applies a further correction from the default.

Change-Id: Ia94e08bc24c67b604831f980909af7e982fcd16d
2013-06-26 17:04:06 +01:00
Jingning Han
d19ea3861d Refactor intra predictor block
Remove vp9_intra4x4_predict(). Use the common intra prediction
function for all block sizes.

Change-Id: Ibd19d51dfa3da8bbdfb79ddeb81530b2e2089560
2013-06-25 16:33:13 -07:00
Dmitry Kovalev
dc0f457c94 Using get_plane_block_{width, height} instead of custom code.
Change-Id: I453ed11b965e857a14c18ea5c0f4a0a48e7dc0d9
2013-06-25 14:11:18 -07:00
Dmitry Kovalev
87ee34aacb Removing unused code.
Removing block index (ib) parameter from get_tx_type_{8x8, 16x16}
functions.

Change-Id: Ia213335aae7a7cb027f97b9cc9b04519840250f1
2013-06-25 10:17:19 -07:00
Dmitry Kovalev
f27f76dfb3 Transforming scale_mv_component_q4 into scale_mv_q4 function.
Using MV instead of int_mv for function arguments.

Change-Id: Ic25e13dccbc98fac1fa1b3255127e00cca2a57f6
2013-06-21 15:34:29 -07:00
Ronald S. Bultje
54b2a59623 Implement SSE2 block_error.
Change vp9_block_error() to return a 64bit error variable, change all
callers to expect a 64bit return value (this will prevent overflows,
which we basically don't check for at all right now). Remove duplicate
block_error() function, which fixed that through truncation. Remove
old (incompatible) mmx/sse2 block_error SIMD versions and replace with
a new one that returns a 64bit value.

Encoding time of first 50 frames of bus @ 1500kbps goes from 3min29 to
3min23, i.e. a 3% overall speedup.

Change-Id: Ib71ac5508b5ee8a80f1753cd85d72df1629abe68
2013-06-21 12:54:52 -07:00
Yaowu Xu
ee07a261a0 rename variables to avoid build error in MSVC
Change-Id: I7960178c95c54d5c4497e44cfc8c493566294b34
2013-06-20 18:31:48 -07:00
Deb Mukherjee
7947a33d72 Improving model rd with variance and quant step
Improves the rd modeling function and implements them using interpolation
from a table which is a little faster. Also uses sse as input to the
modeling function rather than var - since there is no dc prediction
used and as a result the sse works a little better.

derfraw300: +0.05%
Speedup: ~1%

Change-Id: I151353c6451e0e8fe3ae18ab9842f8f67e5151ff
2013-06-20 10:06:28 -07:00
Jim Bankoski
1f94b97694 convert all speed things to speed features
Change-Id: Ie24489a4d39f3e53e816eeebf75a1c9c7d94515a
2013-06-20 09:42:44 -07:00
Yaowu Xu
12180c8329 Remove unnecessary copying of probs.
Change-Id: Ic924f07c6ab0c929c6cdf11880d3c625806e272c
2013-06-18 23:02:27 -07:00
Deb Mukherjee
4ad96115cd Some cleanups in rd motion search
No bitstream or output change - only cosmetics.

Change-Id: Ic8c1d7ad010a87dcf27d12a38cd7dd5adba683a7
2013-06-13 17:25:23 -07:00
Deb Mukherjee
f18328cbf1 Adds a zero check in model_rd function
Avoids divide-by-zero when variance is 0.

Change-Id: I3c7f526979046ff7d17714ce960fe81d6e1442a0
2013-06-10 17:04:47 -07:00
John Koleszar
717d744a01 Fix use of get_uv_tx_size in loopfilter
Change the argument of get_uv_tx_size() to be an MBMI pointer, so that the
correct column's MBMI can be passed to the function.

Change-Id: Ied6b8ec33b77cdd353119e8fd2d157811815fc98
2013-06-10 11:40:57 -07:00
Paul Wilkins
de6ec27d1a Rd check on segment level reference mode.
Do not allow the rd code to check compound modes if
a segment level reference frame is selected.

Change-Id: I95f0c57789e0eaceed7caf227e94b4ba3130a06c
2013-06-10 11:03:15 -07:00
Ronald S. Bultje
b12a8dac98 Allow non-zeromv if ref_frame=intra with segmentation skip/ref enabled.
Change-Id: Ib5a95bb6ab643b276df3faa9bf99595e4a69ff18
2013-06-10 10:55:10 -07:00
Tero Rintaluoma
86bb6df005 Fixed point reference picture scaling
Fixed point scaling factors are calculated once for each
reference frame by using integer division. Otherwise fixed point
scaling routines are used in all scaling calculations. This makes it
possible to calculate fixed point scaling factors on device driver
software and pass them to hardware and thus avoid division on hardware.

TODO:
 - Missing check for maximum frame dimensions
   (currently scaling uses 14 bits)
 - Missing check for maximum scaling ratio
   (upscaling 16:1, downscaling 2:1)

Problems:
 - Straightforward fixed point implementation can cause error +-1
   compared to integer division (i.e. in x_step_q4). Should only
   be an issue for frames larger than 16k.

Change-Id: I3cf4dabd610a4dc18da3bdb31ae244ebaf5d579c
2013-06-10 08:07:55 -07:00
Deb Mukherjee
21401942b0 Coding tx-size selection by use of spatial context
Adds coding of transform size within a frame by use of context
of transform sizes selected in left and above blocks.

Also incorporates code for generating stats.

TODO: generate and incorporate new default stats

Change-Id: I6a7af099f6ad61d448521d9a51167aedaf638ed6
2013-06-07 16:07:58 -07:00
Paul Wilkins
340c7a48e6 Change to segment ref frame feature.
Simplify feature to only support a single reference frame
instead of a mask.

Change-Id: I5dd3a98c7a224aafb35708850ab82e2f220e68fb
2013-06-07 21:42:22 +01:00
Deb Mukherjee
3ee1a21a42 Coding updates for tx-size selection
Changes to the coding of transform sizes, along with forward
and backward probability updates.

Results:
derf300: +0.241%

Context based coding of transform sizes will be in a separate
patch.

Change-Id: I97241d60a926f014fee2de21fa4446ca56495756
2013-06-07 08:54:00 -07:00
Ronald S. Bultje
6ef805eb9d Change ref frame coding.
Code intra/inter, then comp/single, then the ref frame selection.
Use contextualization for all steps. Don't code two past frames
in comp pred mode.

Change-Id: I4639a78cd5cccb283023265dbcc07898c3e7cf95
2013-06-06 17:28:09 -07:00
Ronald S. Bultje
ad34368786 New intra mode and partitioning probabilities.
Split partition probabilities between keyframes and non-keyframes,
since they are fairly different. Also have per-blocksize interframe
y intramode probabilities, since these vary heavily between different
blocksizes.

Lastly, replace default probabilities for partitioning and intra modes
with new ones generated from current codec. Replace counts with actual
probabilities also.

Change-Id: I77ca996e25e4a28e03bdbc542f27a3e64ca1234f
2013-06-06 10:45:30 -07:00
Jingning Han
d03e974fbd Bug fix in rd_pick_inter_mode_sb_
Fix the calculation of step size in height.

Change-Id: I0e0c0175f141f5a41214ae51cef233d13942d3c5
2013-06-06 10:04:26 -07:00
Paul Wilkins
26e24b1dd7 Merge "Rd thresholds change with block size." into experimental 2013-06-06 09:27:44 -07:00
Paul Wilkins
02590a5b1b Merge "Turn off compound inter search refinement for good quality." into experimental 2013-06-06 09:27:31 -07:00
Jim Bankoski
b4c4f64862 signs reverted
Change-Id: Ieface458c83eb6e7ee95595d9fc662f372117c9a
2013-06-06 08:59:22 -07:00
Paul Wilkins
c3316c2bc5 Rd thresholds change with block size.
Added structures to support independent rd thresholds
for different block sizes (and set experimental block
size correction factors).

Added structure to to allow dynamic adaptation of thresholds
per mode and per block size basis depending on how often
the mode/block size combination is seen (currently fixed factor).

Removed some unused variables.

TODO
- Adaptation of thresholds based on how often each mode chosen.
- The baseline mode values could also be adjusted based on
  the block size (e.g. for a particular intra mode use a low threshold
  for 4x4 prediction blocks but a relatively high value for 64x64.

Change-Id: Iddee65ff3324ee309815ae7c1c5a8584720e7568
2013-06-06 15:45:53 +01:00
Paul Wilkins
c880e02f97 Turn off compound inter search refinement for good quality.
Turn this feature off for some modes in  "good" quality.

Change-Id: I3f262d62cca8f01736b977af1465291e8be29f0a
2013-06-06 15:44:25 +01:00
Jim Bankoski
5a88271b09 don't tokenize & encode tokens for blocks in UMV
This avoids encoding tokens for blocks that are entirely
in the UMV border. This changes the bitstream.

Change-Id: I32b4df46ac8a990d0c37cee92fd34f8ddd4fb6c9
2013-06-06 06:10:25 -07:00
Jingning Han
61e6586230 Merge "Fix UV intra coding rd loop" into experimental 2013-06-05 21:47:00 -07:00
Jingning Han
f04b15486a Fix UV intra coding rd loop
This commit makes the coding/reconstruction operations of intra
coding rate-distortion loop for UV components consistent with those
of the encoding process.

key frame coding gains:
derf:   0.11%
stdhd:  0.42%

Change-Id: I8d49f83924a320e3689ef2d60096c49d7f0c7a40
2013-06-05 21:18:02 -07:00
Deb Mukherjee
30226a658f Cosmetic renaming VP9_MVREFS to VP9_INTER_MODES
NO bitstream change

Change-Id: I79f6146dac5fdd157051b6f8dc611c0b7b5e5f7f
2013-06-05 11:24:01 -07:00
Jingning Han
513d326d75 Merge "Make sb intra rd search consistent with encoding" into experimental 2013-06-04 14:59:05 -07:00
Jingning Han
51b6e73a68 Make sb intra rd search consistent with encoding
This commit makes operations of the superblock intra coding rate
distortion optimization consistent with those used in the encoding
process. Given the test prediction mode and transform size, the rd
optimizer encodes and reconstructs each transformed block of the
superblock consecutively, then computes the total rate-distortion
costs accosicated with the current superblock to select the coding
decisions.

It achieves coding performance gains:
derf 0.353%
yt   1.111%

Change-Id: I0da2eb7a71361dfb8c1384927fc536b0c2790d07
2013-06-04 13:54:48 -07:00
Dmitry Kovalev
6a961e7dc8 Merge "Replacing memcpy with struct assignment." into experimental 2013-06-03 14:32:05 -07:00
Jingning Han
9068bce4e7 Put iterative motion search under speed control
Enable iterative motion search for compound inter-inter prediction
of block sizes 4x4/4x8/8x4 only when best coding quality is selected.
The iterative motion search provides about 0.1% gains for derf and
stdhd at this point, at the expense of longer runtime.

Change-Id: Idc03e7f827e51f1bb8d269bc3752ee297a6bbfe5
2013-06-03 09:18:57 -07:00
Dmitry Kovalev
3b9ec31eaf Replacing memcpy with struct assignment.
Change-Id: Ib557cc6351404b9e178e95a545883eb3666f11f0
2013-05-31 16:00:32 -07:00
Dmitry Kovalev
317d832d38 Merge "Adding plane_block_width and plane_block_height functions." into experimental 2013-05-31 15:28:45 -07:00
Deb Mukherjee
0048ec2329 Costing fixes related to trellis optimization
Migrates costing changes/fixes from the rebalance expt to the head
without the expt on.

Rebased.

Change-Id: I51677d62f77ed08aca8d21a4c9a13103eb8de93f
Results:
derfraw300: +0.126%
2013-05-31 13:56:32 -07:00
Dmitry Kovalev
120a878199 Adding plane_block_width and plane_block_height functions.
Change-Id: I02c17fb733c0f3c22dc3167c3d3182797415f1ae
2013-05-31 12:31:49 -07:00
Ronald S. Bultje
a288cb3b10 Merge "Merge all various transform size data trackers into single variables." into experimental 2013-05-31 09:59:24 -07:00
Scott LaVarnway
1e025dbfd1 Merge "Moved use_prev_in_find_mv_refs check to frame level" into experimental 2013-05-31 09:35:51 -07:00
Ronald S. Bultje
e9d68a5e36 Merge all various transform size data trackers into single variables.
Change-Id: I2dfc569106b29fbe4da20585a0e85e5e9ea6a4db
2013-05-31 09:18:59 -07:00
Jim Bankoski
21595f8e38 Merge "Creates a new speed 1:" into experimental 2013-05-30 20:36:05 -07:00
Jim Bankoski
ced21bd6a6 Creates a new speed 1:
This speed 1 - uses variance threshold stolen from static-thresh
to determine split.  Any superblock with greater than the variance
set by static thresh * quantizer index squared is split. In addition
transform size is set to largest size less than or equal to partition
size, sub pixel filter is set to normal,  and only 12 modes are used
at all.

Change-Id: If7a2858ee70f96d1eb989c04fd87a332b147abef
2013-05-30 19:53:00 -07:00
Ronald S. Bultje
16482bddf7 Merge "Remove splitmv." into experimental 2013-05-30 19:07:12 -07:00
Ronald S. Bultje
d2205f92c3 Merge changes I98c18fe5,I80c37cff into experimental
* changes:
  Remove i4x4_pred.
  Remove unused table.
2013-05-30 19:06:44 -07:00
Ronald S. Bultje
e6485581fe Remove splitmv.
We leave it in rdopt.c as a local define for now - this can be removed
later. In all other places, we remove it, thereby slightly decreasing
the size of some arrays in the bitstream.

Change-Id: Ic2a9beb97a4eda0b086f62c039d994b192f99ca5
2013-05-30 17:21:01 -07:00
Ronald S. Bultje
1efa79d32f Remove i4x4_pred.
It remains as a local define in rdopt.c so we can distinguish between
split and non-split modes in the RD loop, but disappears outside that
scope in the codec.

Change-Id: I98c18fe5ab7e4fbd1d6620ec5695e2ea20513ce9
2013-05-30 16:44:58 -07:00
Ronald S. Bultje
f5827699bf Merge "Merge all intra mode coding trees into a single one." into experimental 2013-05-30 11:27:51 -07:00
Jingning Han
5e97862a71 Merge "Enable iterative motion search for 4x4 inter pred" into experimental 2013-05-30 11:02:10 -07:00
Ronald S. Bultje
98c192ae83 Merge all intra mode coding trees into a single one.
Also merge all counters. This removes a few unused probability updates
from the bitstream.

Change-Id: I20f58853e9dac84d8c0d9703ae012c55917516eb
2013-05-30 09:58:53 -07:00
Jim Bankoski
e987f03acd Merge "valgrind - txfm_thresh not set" into experimental 2013-05-30 09:34:48 -07:00
Deb Mukherjee
c98bfcfbbb Merge "Balancing coef-tree to reduce bool decodes" into experimental 2013-05-30 08:10:47 -07:00
Jim Bankoski
ecf023f6e4 Merge "fix valgrind warning" into experimental 2013-05-30 08:04:49 -07:00
Jingning Han
87626a8f6e Enable iterative motion search for 4x4 inter pred
This commit enables iterative motion search for 4x4/4x8/8x4 block
size compound inter-inter prediction.

WIP: borg run testing

Change-Id: I2b318db4a03cdca5a8002b3fa6c0fa89b129288b
2013-05-30 10:49:35 +01:00
Ronald S. Bultje
17544d1478 Merge "Remove some unused code related to macroblock/splitmv coding." into experimental 2013-05-29 17:35:05 -07:00
Jingning Han
5c05fbf6bb Merge "Refactor 4x4 block level rd loop" into experimental 2013-05-29 16:35:02 -07:00
Deb Mukherjee
b8b3f1a46d Balancing coef-tree to reduce bool decodes
This patch changes the coefficient tree to move the EOB to below
the ZERO node in order to save number of bool decodes.

The advantages of moving EOB one step down as opposed to two steps down
in the other parallel patch are: 1. The coef modeling based on
the One-node becomes independent of the tree structure above it, and
2. Fewer conext/counter increases are needed.

The drawback is that the potential savings in bool decodes will be
less, but assuming that 0s are much more predominant than 1's the
potential savings is still likely to be substantial.

Results on derf300: -0.237%

Change-Id: Ie784be13dc98291306b338e8228703a4c2ea2242
2013-05-29 16:25:52 -07:00
Jim Bankoski
aae78c8ac7 valgrind - txfm_thresh not set
For 4x4 blocks valgrind points out the cache was uninitalized.
This resolves the issue by setting it.

Change-Id: I22733000da048643762813a84fbda66d8e4040d2
2013-05-29 13:56:08 -07:00
Jingning Han
d0a3872019 Refactor 4x4 block level rd loop
This commit makes clean-ups in the rate-distortion loop for 4x4,
4x8, and 8x4 block sizes for the use of iterative motion search.

Removed unnecessary use of bmi in handle_inter_mode.

Deprecated loop over labels in the 4x4/4x8/8x4 block rd search.

Change-Id: I71203dbb68b65e66f073b37abd90d82ef5ae6826
2013-05-29 13:44:52 -07:00
Scott LaVarnway
353642bc53 Moved use_prev_in_find_mv_refs check to frame level
This patch checks at the frame level to see if the previous
mode info context can be used.  This patch eliminates the
flag check that was done for every mode and removes another
check that was done prior to every vp9_find_mv_refs().

Change-Id: I9da5e18b7e7e28f8b1f90d527cad087073df2d73
2013-05-29 16:42:23 -04:00
Jim Bankoski
5e5470b254 fix valgrind warning
scales for second reference frame vars are unitialized if the
second ref frame is one of of those disallowed by refframeflags

Change-Id: I4ce42de391178c1699dcaede18c5f12c84993c61
2013-05-29 12:34:10 -07:00
Jingning Han
84deeddbaf Merge "Refactor rd loop for inter modes" into experimental 2013-05-29 10:55:23 -07:00
Jingning Han
6c97bba403 Merge "further clean-ups on intra4x4 coding" into experimental 2013-05-29 10:55:14 -07:00
Sami Pietila
88a4d4c510 Residual coding to cache energy class of tokens.
Proposal for tuning the residual coding by changing how the context
from previous tokens is calculated. Storing the energy class of previous
tokens instead of the token itself eases the critical path of
HW implementations.

Change-Id: I6d71d856b84518f6c88de771ddd818436f794bab
2013-05-29 15:21:01 +01:00
Ronald S. Bultje
4487f5a690 Remove some unused code related to macroblock/splitmv coding.
Change-Id: Ic40d56fb162f4e201547dfae33e62ccd9e865889
2013-05-29 06:29:56 -07:00
Jingning Han
94d700e763 Refactor rd loop for inter modes
This commit pulls the iterative motion search for compound inter-
inter out from handle_inter_mode_ as a separate function. Hence,
it is applicable to 4x4/4x8/8x4 level compound inter search to be
enabled later.

Also edit the rd loop for 4x4 inter block sizes for cosmetic
purpose.

Change-Id: Ibc71a11cbe5a26cd52faba01026cf8446cf4d2b4
2013-05-28 16:31:33 -07:00
Jingning Han
4729a6f389 further clean-ups on intra4x4 coding
Removed one 4x4 prediction step that was unnessary in the rd loop.
Removed a unused modecosts estimate from encoder side.

Change-Id: I65221a52719d6876492996955ef04142d2752d86
2013-05-28 11:19:05 -07:00
Yaowu Xu
601bab4fde Merge "a few clean-ups" into experimental 2013-05-27 15:16:21 -07:00
Ronald S. Bultje
cba8e16e93 Decrease scope of frame_mv argument to handle_inter_mode().
Change-Id: I81c637c61ecc33cb66beb59a2a33166d66b9a0a2
2013-05-27 14:16:45 -07:00
Yaowu Xu
2b96ffe025 a few clean-ups
1. remove prediction mode conversion
2. unified bmode, same for key and non-key frame
3. set I4X4_PRED count for pdf to 0, as I4X4_PRED is no longer
coded ever. It is determined by ref_frame and block partition

Change-Id: If5b282957c24339b241acdb9f2afef85658fe47d
2013-05-27 13:53:56 -07:00
Ronald S. Bultje
f188bf1c3d Remove unused mode_index argument from handle_inter_mode().
Change-Id: I07b8c15f33e6e7c63dd0033c18c4ac5c0303cf32
2013-05-27 08:49:17 -07:00
Ronald S. Bultje
5cac66078e Remove splitmv.
Also do per-partition motion vector referencing in <sb8x8 partitions,
and adjust mvref finding for sub8x8 partitions.

Change-Id: Id3ed1ed4d2a8910d11d327db6cc63b8eb79f941f
2013-05-26 14:40:49 -07:00
Jingning Han
826efc838c Fix a bug in intra4x4 level rd loop
This commit fixed a uninitialized value use in the intra 4x4/8x4/4x8
rate-distortion loop.

Change-Id: I5c25b3536b59e4f5fbb23cf85baf93b2ccec7d72
2013-05-23 17:44:33 -07:00
Jingning Han
ae10319520 Make comp_inter_inter support 4x4 partition coding
This commit refactors the iterative motion search for compound
inter-inter mode, to make it support all partition types including
4x4/4x8/8x4 block sizes.

Change-Id: I5f1212b0f307377291763e45c6bdc9693b5f04c8
2013-05-23 13:13:42 +01:00
Paul Wilkins
33ecd6ad54 Merge Scatter Scan experiment.
Removal from under configure flag.
A bit  renaming

Change-Id: I2213229dfe852001dfec16b149f47c52ce88f3aa
2013-05-23 13:09:27 +01:00
Jingning Han
7ac5ac52f9 Merge 4x4 block level partition into codebase
Move 4x4/4x8/8x4 partition coding out of experimental list.

This commit fixed the unit test failure issues. It also resolved
the merge conflicts between 4x4 block level partition and iterative
motion search for comp_inter_inter.

Change-Id: I898671f0631f5ddc4f5cc68d4c62ead7de9c5a58
2013-05-23 11:58:50 +01:00
Deb Mukherjee
ddb2309568 Merge "Using 128 entry look up table for coef models" into experimental 2013-05-22 10:38:35 -07:00
Jingning Han
d2cacdc530 Merge "Make the intra rd search support 8x4/4x8" into experimental 2013-05-22 10:00:15 -07:00
Deb Mukherjee
de4d682ca4 Using 128 entry look up table for coef models
Reverts to using 128 bit LUT for the coef models rather than 48
to ease hardware implementation.

Also incorporates some cleanups including removing various
hooks to support different lookup tables based on block_type and
ref_type.

Change-Id: I54100c120cca07a2ebd3a7776bc4630fa6a153f6
2013-05-22 08:44:31 -07:00
Paul Wilkins
0b713f8c18 Merge CONFIG_COMP_INTER_JOINT_SEARCH.
Merge this experiment so that it is under a speed feature
flag not a configuration flag.

Change-Id: I536f7f125a4ff5149bb3a64f791e835c324535fd
2013-05-22 11:23:31 +01:00
Jingning Han
f153a5d063 Make the intra rd search support 8x4/4x8
This commit allows the rate-distortion optimization of intra coding
capable of supporting 8x4 and 4x8 partition settings.

It enables the entropy coding of intra modes in key frame using a
unified contextual probability model conditioned on its above/left
prediction modes.

Coding performance:
derf 0.464%

Change-Id: Ieed055084e11fcb64d5d5faeb0e706d30268ba18
2013-05-21 21:03:00 -07:00
John Koleszar
ddf13be8ef Merge "Initial version of alpha channel support" into experimental 2013-05-21 17:29:51 -07:00
Deb Mukherjee
7a645e4e12 Merging the model coef prob experiment
Merges the experiment.

Change-Id: I4eb19af6de6df6aa3a96a2e82f231d47ed9b3ae9
2013-05-21 14:44:38 -07:00
Scott LaVarnway
1db6373267 Merge "WIP: 4x4 idct/recon merge" into experimental 2013-05-21 10:45:53 -07:00
Dmitry Kovalev
4ac70bd7d3 Adding get_ref_frame_idx function.
Change-Id: I4f1a4eca6794cda78d00512196caacd5567e2dcc
2013-05-20 16:09:00 -07:00
Deb Mukherjee
39a90bc8e8 Updating the model coef experiment
Cleans up the experiment. Actually uses reduced counts for backward
updates, and reduced number of probabilities in the context.

No change in bitstream when the experiment is on.

Between expt on and off:
derfraw300 is down only -0.062% (which is better than when expts
were run previously).

Change-Id: I55285a049a0c22810bdb42914212ab5a4f8521b5
2013-05-20 12:46:36 -07:00
Scott LaVarnway
ba48a11130 WIP: 4x4 idct/recon merge
This patch eliminates the intermediate diff buffer usage by
combining the short idct and the add residual into one function.
The encoder can use the same code as well.

Change-Id: I296604bf73579c45105de0dd1adbcc91bcc53c22
2013-05-20 13:03:17 -04:00
Jingning Han
810b612c23 Enable bit-stream support to 8x4 and 4x8 partition
The recursive partition type search is enabled down to 4x4, 4x8 and
8x4, followed by the corresponding rate-distortion optimization for
the per-partition encoding mode decisions.

The bit-stream writing/reading synchronized in supporting the
rectangular partition of 8x8 block.

This provides above 1% coding performance gains on derf.

To do next:
1. re-design the rate-distortion loop for inter prediction below 8x8.
2. re-design the rate-distortion loop for intra prediction below 4x4.
3. make the loop-filter aware of rectangular partition of 8x8 block.
4. clean the unused probability models.
5. update default probability values.

Change-Id: Idd41a315b16879db08f045a322241f46f1d53f20
2013-05-19 14:59:04 -07:00
John Koleszar
679e4abdd5 Initial version of alpha channel support
This is a mostly-working implementation of an extra channel in the
bitstream. Configure with --enable-alpha to test. Notable TODOs:

 - Add extra channel to all mismatch tests, PSNR, SSIM, etc
 - Configurable subsampling
 - Variable number of planes (currently always uses all 4)
 - Loop filtering
 - Per-plane lossless quantizer
 - ARNR support

This implementation just uses the same contents as the Y channel
for the A channel, due to lack of content and general pain in
playing back 4 channel content. A later patch will use the actual
alpha channel passed in from outside the codec.

Change-Id: Ibf81f023b1c570bd84b3064e9b4b8ae52e087592
2013-05-16 22:21:09 -07:00
Jingning Han
8e3d0e4d7d Add building blocks for 4x8/8x4 rd search
These building blocks enable rate-distortion optimization search
over block sizes of 8x4 and 4x8. Need to convert them into mmx/sse
forms.

Change-Id: I570ea2d22d14ceec3fe3575128d7dfa172a577de
2013-05-16 10:41:29 -07:00
Jingning Han
8468a5c1a0 Fix the transform type selection in 4x4 partition
This commit allows proper transform type (DCT/ADST) selection in
the settings of partition 4x4 level.

Change-Id: Iec6f922a46480d777e7ca9142a99e8c131f0077b
2013-05-15 16:09:58 -07:00
Jingning Han
1f26840fbf Enable recursive partition down to 4x4
This commit allows the rate-distortion optimization recursion
at encoder to go down to 4x4 block size. It deprecates the use
of I4X4_PRED and SPLITMV syntax elements from bit-stream
writing/reading. Will remove the unused probability models in
the next patch.

The partition type search and bit-stream are now capable of
supporting the rectangular partition of 8x8 block, i.e., 8x4
and 4x8. Need to revise the rate-distortion parts to get these
two partition tested in the rd loop.

Change-Id: I0dfe3b90a1507ad6138db10cc58e6e237a06a9d6
2013-05-14 12:39:56 -07:00
Yunqing Wang
dee12bdf8f Merge "Do joint motion search iteratively" into experimental 2013-05-14 10:18:11 -07:00
Yunqing Wang
60456083e9 Do joint motion search iteratively
Allow motion search multiple times iteratively, and break out
the loop if this search couldn't find better motion vectors.
Limit the maximum number of search to 2.

Tests results:
1. stdhd set: 0.311%(overall psnr); 0.346%(ssim).
positive gain on 10 out of 16 clips(best: 2.746% on sunflower;
worst: -0.434% on old_town_cross).
2. derf set: 0.016%(overall psnr); 0.062%(ssim).
positive gain on half of the clips(best: 0.499% on bowing;
worst: -0.387 on city).

Change-Id: Ibf0a51776d4caf7707be0586346db08128117559
2013-05-13 12:14:09 -07:00
Jingning Han
e996c9c5f1 Merge "Force bsize for UV in I4X4 and SPLITMV" into experimental 2013-05-13 10:51:39 -07:00
Paul Wilkins
e5f715201a Change to band calculation.
Change band calculation back to simpler model based
on the order in which coefficients are coded in scan order
not the absolute coefficient positions.

With the scatter scan experiment enabled the results were
appear broadly neutral on derf (-0.028) but up a little on std-hd +0.134).

Without the scatterscan experiment on the results were up derf as well.

Change-Id: Ie9ef03ce42a6b24b849a4bebe950d4a5dffa6791
2013-05-13 17:21:49 +01:00
Jingning Han
4c2c350309 Force bsize for UV in I4X4 and SPLITMV
Use 4x4 block coding for UV components arbitrarily in I4X4_PRED and
SPLITMV coding modes. This is a temporary solution to enable
bit-stream support for recursive partition down to 4x4 block size.
Will separate the functionalities of 4x4 block coding rate-distortion
out from those of superblocks.

Change-Id: I03dc15d5897014f175f3f2c91e9b266091d56797
2013-05-11 13:39:16 -07:00
Yunqing Wang
9755d9fda2 Remove unused mdcounts
mdcounts seems no longer used.

Change-Id: Idd8162e8acfa3f5be7a18767156cc79ccbc2bdee
2013-05-10 11:02:22 -07:00
Yunqing Wang
9f5811c2da Add joint motion search in comp_inter_inter mode(experiment)
In current code, motion vectors got from single prediction mode are used
in compound prediction mode directly. These motion vectors may not give
accurate prediction since they are searched independently. In this patch,
we took Pascal's suggestion, and did joint motion search in compound
prediction mode to find better motion vectors in this situation.
Test results:
Overall PSNR: 0.570%(derf), 0.918%(stdhd);
SSIM: 0.572%(derf), 1.009%(stdhd);

The encoder is a little slower. This can be improved since some c
code is used in motion search.

Change-Id: Ib30c9240f6c56c9b070867b4ca89412a76d9f3c6
2013-05-10 10:15:43 -07:00
Dmitry Kovalev
f0911886f3 Merge "Renaming 'Speed' to 'speed' inside VP9_COMP struct." into experimental 2013-05-08 16:35:35 -07:00
Dmitry Kovalev
8f4e9ac8bc Removing y_to_uv_block_size and y_bsizet_to_block_size functions.
Change-Id: I49527ff8dd8bef1074c18a964fed2a575f0b118a
2013-05-08 15:23:42 -07:00
Dmitry Kovalev
4be190d9d0 Renaming 'Speed' to 'speed' inside VP9_COMP struct.
Change-Id: I4374b5af40ee9082ddf7956a9756a15ad9ad5436
2013-05-08 14:35:42 -07:00
John Koleszar
14a5c7285b Make switchable filter search subsampling-aware
Makes the temporary storage of the filtered data agnostic to
the number of planes and how they're subsampled.

Change-Id: I12f352cd69a47ebe1ac622af30db29b49becb7f4
2013-05-07 21:57:00 -07:00
John Koleszar
7465f52f81 Merge "Make setup_pred_block subsampling-aware." into experimental 2013-05-07 21:53:31 -07:00
Dmitry Kovalev
80997b3aa2 Merge "Adding get_switchable_rate function." into experimental 2013-05-07 17:10:48 -07:00
Paul Wilkins
a14ae84749 Deprecate code_zerogroup experiment.
Delete code under the CONFIG_CODE_ZEROGROUP flag.

Change-Id: I5fe6c7b42a5da9b73118e33594301da4129f320a
2013-05-07 16:52:55 -07:00
Dmitry Kovalev
455816231e Adding get_switchable_rate function.
Change-Id: I71311a14f8d7f48508b250f25d1d0914c6a1ac72
2013-05-07 16:52:04 -07:00
Paul Wilkins
1ed57a6a62 Deprecate comp_interintra_pred experiment.
Delete code under the CONFIG_COMP_INTERINTRA_PRED
flag.

Change-Id: I3d1079cf46305c08f7e11d738596ea112e7b547f
2013-05-07 16:24:08 -07:00
Paul Wilkins
8c1b516d10 Deprecate the newbintramode experiment.
Clean out code relating to newbintramode.

Change-Id: Ie91f4f156cdf60ce0da8ca407c1c9cb00c7d0705
2013-05-07 16:00:59 -07:00
Jingning Han
cf8b5a09ed Add building blocks for partition down to 4x4
Macro ab4x4 contains experiments for recursive partition down to
4x4 block size.

Change-Id: Ic727842fa98a4df9fd51e0025a545dc76a5c76c1
2013-05-07 12:11:51 -07:00
John Koleszar
e559e14fa6 Make setup_pred_block subsampling-aware.
Code previously set up the pointers by scaling by MI_UV_SIZE, which
is 4:2:0 only.

Change-Id: Ic13a92895cff018ec1345736746ed84cb31e6e31
2013-05-07 11:47:45 -07:00
Jingning Han
776c1482a3 Merge SB8X8 into the codebase
Pull sb8x8 out of experimental list. verified via borg run tests.
Fixed unit test failures.

Change-Id: I12a4bbd17395930580c048ab68becad1ffe46e76
2013-05-07 09:08:25 -07:00
Dmitry Kovalev
2e5f0084f3 Adding model_rd_for_sb function.
Iterating over all planes in the loop instead of custom y,uv code inside
handle_inter_mode function.

Change-Id: I301f9276d6d544c2fd7203d84f1318ac80ea625d
2013-05-06 12:42:53 -07:00
Jingning Han
8e1c97cf73 Fix a unit test failure of sb8x8 on scaling ref
Disable the use of scaled reference frame for motion search in
SPLITMV mode. This fixes the unit test failure issue triggered
when merging sb8x8 from experimental list.

Change-Id: I02ac25fd8db8d5762f8fee29513b947189875fa0
2013-05-06 10:28:18 -07:00
Ronald S. Bultje
f7fa367094 Fix first-pass intra4x4 for sb8x8 experiment.
Change-Id: I1df17f45721c690d157800daa6a0b377e3d32bc2
2013-05-04 15:49:41 -07:00
Ronald S. Bultje
842c573e04 Merge "Fix overflow in RD error calculation code." into experimental 2013-05-03 18:03:06 -07:00
John Koleszar
6c622e2783 Merge "Separate transform and quant from vp9_encode_sb" into experimental 2013-05-03 17:19:01 -07:00
John Koleszar
4529c68b3b Separate transform and quant from vp9_encode_sb
This allows removing a large number of transform size specific functions,
as well as supporting 444/alpha by routing all code through the
subsampling-aware path.

Change-Id: Ieb085cebe9f37f24fc24de179898b22abfda08a4
2013-05-03 12:14:50 -07:00
Ronald S. Bultje
ee808e52bd Fix overflow in RD error calculation code.
Change-Id: I61ef1f198c876f9f79787ea7d7385a862cfbae19
2013-05-03 10:33:07 -07:00
Dmitry Kovalev
7ab2d7bf55 Removing MAXF macro and using MAX instead.
Change-Id: I51c53692b1150005645bf362c5e5a8275178a8fd
2013-05-02 11:57:16 -07:00
Ronald S. Bultje
f37d8400db Store splitmv modes in context after 8x8 rd loop.
Change-Id: I07aa89a67e0ac5f99ef0c448553dbc46b0ed27f2
2013-05-01 17:13:23 -07:00
Ronald S. Bultje
b6c2d872f0 Fix some crashes in sb8x8 experiment.
Change-Id: I390bb1cedc835f439fd5dd6cda6572b29cbb139c
2013-05-01 14:45:27 -07:00
Dmitry Kovalev
79590f186c Merge "Cleaning up encoder segmentation code." into experimental 2013-04-30 17:49:55 -07:00
Ronald S. Bultje
d068d869b9 sb8x8 integration in rd loop.
Work-in-progress, not yet ready for review. TODO items:
- bitstream writing (encoder) and reading (decoder)
- decoder reconstruction

Change-Id: I5afb7284e7e0480847b47cd0097cb469433c9081
2013-04-30 16:13:20 -07:00
Dmitry Kovalev
51a73fbba2 Merge "Consistent names for quant-related functions and variables." into experimental 2013-04-30 10:19:48 -07:00
Dmitry Kovalev
ee97da2c03 Cleaning up encoder segmentation code.
Moving code from vp9_pack_bitstream to new function encode_segmentation.

Change-Id: I1f1e59a1f038618ad95162b7db4b6f8164850ea8
2013-04-29 16:07:17 -07:00
Ronald S. Bultje
2dbaa4f4f4 Change above/left_context to use an 8x8 basis.
Output changes slightly because of a minor bug in (at least) the sb32x16
block2above tx16x16 tables that previously existed in vp9_blockd.c.

Change-Id: I624af28ac200a8322d64454cf05c79e9502968cc
2013-04-29 10:37:25 -07:00
Dmitry Kovalev
5a5a1f25a8 Consistent names for quant-related functions and variables.
Change-Id: I3a6d601e90e8740b9c26dd0afbfe9d467b75d367
2013-04-26 12:30:20 -07:00
Ronald S. Bultje
1a46b30ebe Grow MODE_INFO array to use an 8x8 basis.
Change-Id: I087e08e7909a406b71715b8525c104208daa6889
2013-04-26 11:57:17 -07:00
John Koleszar
bb41ab4a0c Remove BLOCKD structure
All members can be referenced from their per-plane counterparts, and
removes assumptions about 24 blocks per macroblock.

Change-Id: I7ff2fa72d22c29163eb558981c8193765a8113d9
2013-04-26 10:35:54 -07:00
John Koleszar
4f55c5618a Remove destination pointers from BLOCKD
Access these members from MACROBLOCKD instead.

Change-Id: I7907230dd473ff12ebe182b9280d8b7f12a888c4
2013-04-26 10:14:07 -07:00
John Koleszar
4b27eb1f18 Merge "quantize: make 4x4, 8x8 common with larger transforms" into experimental 2013-04-26 09:08:49 -07:00
Scott LaVarnway
57f180b388 Removed bmi from blockd
This originally was "Removed update_blockd_bmi()".  Now,
this patch removed bmi from blockd and uses the bmi found
in mode_info_context.  Eliminates unnecessary bmi copies between
blockd and mode_info_context.

Change-Id: I287a4972974bb363f49e528daa9b2a2293f4bc76
2013-04-26 10:19:43 -04:00
John Koleszar
a672351af9 quantize: make 4x4, 8x8 common with larger transforms
There were 4 variants of the quantize loop in vp9_quantize.c, now
there is 1.

Change-Id: Ic853393411214b32d46a6ba53769413bd14e1cac
2013-04-25 14:44:54 -07:00
Ronald S. Bultje
18f29ff581 Remove duplicate code in RD handle_inter_mode() function.
Change-Id: I552d53f7e7331e9246d8a32d6c6dcc0cfa0cbeb0
2013-04-25 14:21:21 -07:00
Ronald S. Bultje
c849eaca59 Use b_width/height_log2 instead of mb_ where appropriate.
Basic assumption: when talking about transform units, use b_; when
talking about macroblock indices, use mb_.

Change-Id: Ifd163f595d4924ff892de4eb0401ccd56dc81884
2013-04-25 14:20:59 -07:00
John Koleszar
a99e1aa8ca Remove predictor pointers from BLOCKD
Access these members from MACROBLOCKD instead.

Change-Id: I2574622e577bb9feede47f6b7ccbb11f3e928ca8
2013-04-25 12:04:07 -07:00
John Koleszar
6c0c6b86c1 Remove diff from BLOCKD
The underlying storage for these buffers is in the per-plane MACROBLOCKD
area, so read it from there directly.

Change-Id: Id6bd835117fdd9dea07db95ad06eff9f12afaaf7
2013-04-25 11:57:22 -07:00
John Koleszar
15255eef82 Move dequant from BLOCKD to per-plane MACROBLOCKD
This data can vary per-plane, but not per-block.

Change-Id: I1971b0b2c2e697d2118e38b54ef446e52f63c65a
2013-04-25 11:57:20 -07:00
John Koleszar
4bd0f4f646 Remove BLOCK structure
All members can be referenced from their per-plane counterparts, and
removes assumptions about 24 blocks per macroblock.

Change-Id: I593fb0715e74cd84b48facd1c9b18c3ae1185d4b
2013-04-25 11:33:17 -07:00
Dmitry Kovalev
61a47da869 Adding is_inter_mode function.
Change-Id: I2d32d46002cb92c63050c2b8328865c406103621
2013-04-25 10:23:00 -07:00
Jingning Han
b0e3b3df18 Move sbsegment out of experimental list
Move rectangular superblock coding out of experimental list.

Change-Id: I96c37547d122330d666a67b4bf577ae54547857f
2013-04-24 15:19:17 -07:00
Jingning Han
ff2b8aa2c9 Contextual entropy coding of partition syntax
This commit enables selecting probability models for recursive block
partition information syntax, depending on its above/left partition
information, as well as the current block size. These conditional
probability models are reasonably stationary and consistent across
frames, hence the backward adaptive approach is used to maintain and
update the contextual models.

It achieves coding performance gains (on top of enabling rectangular
block sizes):
derf:   0.242%
yt:     0.391%
hd:     0.376%
stdhd:  0.645%

Change-Id: Ie513d9673337f0d27abd65fb566b711d0844ec2e
2013-04-24 14:23:14 -07:00
John Koleszar
bc30736f9b Merge "Remove coeff from BLOCK" into experimental 2013-04-23 17:42:12 -07:00
John Koleszar
aa6a36b062 Merge "Convert coeff to per-plane MACROBLOCK data" into experimental 2013-04-23 17:41:59 -07:00
John Koleszar
48f3e66e16 Remove coeff from BLOCK
Lookup the data per-plane from the MACROBLOCK struct.

Change-Id: I9253c4d3cf886aa9ab4aeab23a2156bfcf994ede
2013-04-23 16:39:21 -07:00
John Koleszar
138ec38cab Convert coeff to per-plane MACROBLOCK data
This commit moves the coeff storage from the MACROBLOCK struct to its
per-plane part. The next commit will remove the coeff member from the
BLOCK structure so that it is consistently accessed per-plane.

Also refactors vp9_sb_block_error_c and vp9_sb_uv_block_error_c to be
variable subsampling aware.

Change-Id: I18c30f87f27c3a012119b6c1970d5fa499804455
2013-04-23 16:28:17 -07:00
John Koleszar
4f35e3e1c1 Merge "Move src_diff to per-plane MACROBLOCK data" into experimental 2013-04-23 16:24:08 -07:00
Dmitry Kovalev
d0d1094a05 Merge "Adding get_scan_{4x4, 8x8, 16x16} functions." into experimental 2013-04-23 12:44:51 -07:00
John Koleszar
cbd1315ac4 Move src_diff to per-plane MACROBLOCK data
First in a series of commits making certain MACROBLOCK members
addressable per-plane. This commit also refactors the block subtraction
functions vp9_subtract_b, vp9_subtract_sby_c, etc to be
loops-over-planes and variable subsampling aware.

Change-Id: I371d092b914ae0a495dfd852ea1a3d2467be6ec3
2013-04-23 12:18:51 -07:00
Deb Mukherjee
611b26bbe0 Merge "Removing the implicit compound inter experiment" into experimental 2013-04-22 23:22:28 -07:00
Deb Mukherjee
735febf1ce Removing the implicit compound inter experiment
Removing this experiment for now, since it has been broken with
the latest code changes.

Change-Id: I1be2181b56de490fcb577f5905b5e147a8ed82d8
2013-04-22 16:46:54 -07:00
Jim Bankoski
366ff224ef Merge "new version of speed 1" into experimental 2013-04-22 16:42:33 -07:00
Jim Bankoski
e7bddba149 new version of speed 1
This version of speed 1 only disables modes at higher resolution that
had distortions >2x the best mode we found...

The hope is that this could be a replacement for speed 0 ...

Change-Id: I7421f1016b8958314469da84c4dccddf25390720
2013-04-22 15:42:41 -07:00
Dmitry Kovalev
5de7e16ca2 Adding get_scan_{4x4, 8x8, 16x16} functions.
Change-Id: Id4306ef6d65d4a3984aed50b775bdf48d4f6c438
2013-04-22 14:08:41 -07:00
John Koleszar
a443447b8b Move pre, second_pre to per-plane MACROBLOCKD data
Continue moving framebuffers to per-plane data.

Change-Id: I237e5a998b364c4ec20316e7249206c0bff8631a
2013-04-22 12:05:24 -07:00
Deb Mukherjee
f12509f640 Merge "Removes the code_nonzerocount experiment" into experimental 2013-04-22 11:53:14 -07:00
Deb Mukherjee
0aa79be7d5 Removes the code_nonzerocount experiment
This patch does not seem to give any benefits.

Change-Id: I9d2b4091d6af3dfc0875f24db86c01e2de57f8db
2013-04-22 10:58:49 -07:00
Deb Mukherjee
6ce718eb18 Merge "End of orientation zero group experiment" into experimental 2013-04-22 10:33:12 -07:00
Deb Mukherjee
70d9f116fd End of orientation zero group experiment
Adds an experiment that codes an end-of-orientation symbol
for every eligible zero encountered in scan order.

This cleans out various other sub-experiments that were part
of the origiinal patch, which will be later included if found
useful.

Results are slightly positive on all sets (0.1 - 0.2% range).

Change-Id: I57765c605fefc7fb9d1b57f1b356843602abefaf
2013-04-22 09:27:59 -07:00
John Koleszar
6d5ac8f2e1 reconinter: remove unnecessary functions, params
Removes the redundant dst pointers from vp9_build_inter_predictors_sb{y,uv}
and the remaining mb specific functions.

Change-Id: I7b6bf439d9394b85ea79b4fe61a3ffc1025720da
2013-04-22 08:20:54 -07:00
John Koleszar
fa8ddbd2a6 Merge "Move dst to per-plane MACROBLOCKD data" into experimental 2013-04-19 16:33:45 -07:00
John Koleszar
d12376aa2c Move dst to per-plane MACROBLOCKD data
First in a series of commits moving the framebuffers pointers to
per-plane data, so that they can be indexed numerically rather than
by name.

Change-Id: I6e0d60fd4d51e6375c384eb7321776564df21775
2013-04-19 16:16:10 -07:00
Yunqing Wang
25edb68100 Merge "Remove unused parameters in handle_inter_mode" into experimental 2013-04-19 14:12:43 -07:00
Paul Wilkins
fb754fd37e Merge "Mv ref candidates cut to 2." into experimental 2013-04-19 14:09:44 -07:00
Dmitry Kovalev
3689122b1c Merge "Fixing member names inside TOKENVALUE and TOKENEXTRA structs." into experimental 2013-04-19 10:09:04 -07:00
Jim Bankoski
35b1d2e38f Merge "catch all for new block sizes" into experimental 2013-04-19 09:57:38 -07:00
Jim Bankoski
afb04eb211 catch all for new block sizes
Just make sure we don't stop them from testing in speed 1.

Change-Id: Iec9b3dba0a32616ff7a451207e0f54b81bb72575
2013-04-19 09:48:56 -07:00
Jim Bankoski
6d82fe219d Merge "set up a new speed 1" into experimental 2013-04-19 08:28:35 -07:00
Paul Wilkins
de80da39dc Mv ref candidates cut to 2.
Further simplification of mvref search to return
only the top two candidates. Distance weights removed
as the test order reflects distance anyway.

Change-Id: I0518cab7280258fec2058670add4f853fab7b855
2013-04-19 16:13:53 +01:00
Jim Bankoski
b6ef0823c5 set up a new speed 1
slightly worse results for faster encodes

Change-Id: I25ea82a18ce20635dbcd328808c1d05ac1f58fd7
2013-04-19 08:04:57 -07:00
Paul Wilkins
92e8a3f514 Simplification of MVref search.
As we are no longer able to sort the candidate
mvrefs in both encoder and decode and given
that the cost of explicit signalling has proved
prohibitive, it no longer makes sense to find more
than 2 candidates.

This patch:

Modifies and simplifies add_candidate_mv()

Removes the forced addition of a 0 vector in the
MAX_MV_REF_CANDIDATES-1 position (in preparation
to reducing MAX_MV_REF_CANDIDATES to 2).

Re-orders the addition of candidates slightly.

This actually gives small gains (circa 0.2% on std-hd)

A subsequent patch will remove NEW_MVREF experiment,
reduce MAX_MV_REF_CANDIDATES to 2 and remove distance
weights as these are implicit now in the order.

Change-Id: I3dbe1a6f8a1a18b3c108257069c22a1141a207a4
2013-04-19 11:19:59 +01:00
Dmitry Kovalev
77f4697a13 Fixing member names inside TOKENVALUE and TOKENEXTRA structs.
Change-Id: I183ec5819d4d80966c92db36db75b8c3be0d381d
2013-04-18 16:18:08 -07:00
Jingning Han
f0b065e946 Merge "Make the use of pred buffers consistent in MB/SB" into experimental 2013-04-18 15:24:55 -07:00
Jingning Han
6f43ff5824 Make the use of pred buffers consistent in MB/SB
Use in-place buffers (dst of MACROBLOCKD) for  macroblock prediction.
This makes the macroblock buffer handling consistent with those of
superblock. Remove predictor buffer MACROBLOCKD.

Change-Id: Id1bcd898961097b1e6230c10f0130753a59fc6df
2013-04-18 14:59:36 -07:00
Dmitry Kovalev
a8d903e539 Merge "Replacing VP9_COMBINEENTROPYCONTEXTS macro with function." into experimental 2013-04-18 14:26:34 -07:00
Dmitry Kovalev
8b20aa2337 Merge "Renaming y1dc_delta_q, uvdc_delta_q, uvac_delta_q fields from VP9Common." into experimental 2013-04-18 14:26:06 -07:00
Yunqing Wang
e304160885 Remove unused parameters in handle_inter_mode
Removed 2 unused parameters.

Change-Id: Ic2862569313c404047072b268c3d2be3f635492c
2013-04-18 11:55:46 -07:00
Ronald S. Bultje
e693472236 Fairly basic integration of rectangular blocks in encoding RD loop.
Adds RD integration for 32x16, 16x32, 64x32 and 32x64 rectangular blocks.
Derf almost +0.6%, HD a little over +1.0%, STDHD +1.3%.

Change-Id: Id651fdb6a655fdbb5c47009757e63317acfb88a5
2013-04-17 09:25:06 -07:00
Dmitry Kovalev
9087d6d470 Replacing VP9_COMBINEENTROPYCONTEXTS macro with function.
Change-Id: I3bbc31840af69481e1d9bb4427c9ee25abf82946
2013-04-16 15:30:28 -07:00
Dmitry Kovalev
1ad7c1f250 Renaming y1dc_delta_q, uvdc_delta_q, uvac_delta_q fields from VP9Common.
New names are y_dc_delta_q, uv_dc_delta_q, uv_ac_delta_q.

Change-Id: I4acae1fc23a4697ce2c5a5becb8dc28ef0a4b552
2013-04-16 15:05:52 -07:00
John Koleszar
e3cfe4e89e Remove the mb_no_coeff_skip flag
This flag was added to VP8 to allow a mode where MB-level skipping
was not allowed, saving a bit per mb. It was never used in practice,
and hasn't been tested in VP9, so remove it.

Change-Id: Id450ec6904c6d06c1919508e7efc52d05cde5631
2013-04-16 12:36:16 -07:00
Dmitry Kovalev
a0d9309eab Removing TRUE and FALSE macro definitions.
Using regular 0 and 1 constants now.

Change-Id: Ie763503cbb727847cc8f1d6506cd6f2ee607f056
2013-04-15 15:24:39 -07:00
Ronald S. Bultje
33a8df085d Fix lingering x->skip settings if static_threshold is used.
Keyframes don't set this variable, so it would use the last set
values from inter frames.

Change-Id: Ie1ef45ece2c44b21b5d55f6cea9f7d6e7a445692
2013-04-15 13:39:07 -07:00
Jingning Han
aaf33d7df5 Add rectangular block size variance/sad functions.
With this, the RD loop properly supports rectangular blocks.

Change-Id: Iece79048fb4e84741ee1ada982da129a7bf00470
2013-04-15 13:39:07 -07:00
Ronald S. Bultje
15eac18c4e Make filter RD code and encode breakout variance size-independent.
Static threshold results slightly up (+0.1% on derf), probably b/c
we now take the filter (sharp/lowpass) into account for the breakout
decision.

Change-Id: I9f597601da434205142afd05f32690e7ba8fd690
2013-04-15 13:38:35 -07:00
Jingning Han
3ba9dd4165 Enable inter predictor for rectangular block size
Combine superblock inter predictors into a unified function that
allows configurable block width and height. The inter predictions
of block sizes smaller than 16x16 are handled differently. To be
continued on merging them later.

Change-Id: I14075959dd5e221f00c205c99ca35c1c31ef728e
2013-04-12 11:51:58 -07:00
Yaowu Xu
7de5edd14a Rename B_PRED to I4X4_PRED
So it is consistent with I8x8_PRED.

Change-Id: Iefa65124b2419690d83e526c611129c0ede29d11
2013-04-12 09:23:58 -07:00
Jingning Han
815e95fbeb Make intra predictor support rectangular blocks
The intra predictor supports configurable block sizes. It can handle
intra prediction down to 4x4 sizes, when enabled in BLOCK_SIZE_TYPE.

Change-Id: I7399ec2512393aa98aadda9813ca0c83e19af854
2013-04-11 16:45:57 -07:00
Scott LaVarnway
cff266bbef Merge "WIP: removing predictor buffer usage from decoder" into experimental 2013-04-11 15:24:33 -07:00
Ronald S. Bultje
69902c6bf0 Merge "Merge pick_sb_modes and pick_sb64_modes." into experimental 2013-04-11 15:06:37 -07:00
Scott LaVarnway
6189f2bcb1 WIP: removing predictor buffer usage from decoder
This patch will use the dest buffer instead of the
predictor buffer.  This will allow us in future commits
to remove the extra mem copy that occurs in the dequant
functions when eob == 0.  We should also be able to remove
extra params that are passed into the dequant functions.

Change-Id: I7241bc1ab797a430418b1f3a95b5476db7455f6a
2013-04-11 13:55:18 -07:00
John Koleszar
c2bd46bf45 tokenize: convert skippable functions
Use the common block walker to calculate skippability.

Change-Id: I6721e42f065df237426c91c1d871ec226ba7cdcb
2013-04-11 12:27:37 -07:00
Ronald S. Bultje
605ff051f7 Merge pick_sb_modes and pick_sb64_modes.
Change-Id: Iad69e7a3b7e470acf6094f6a52e7da69066fd552
2013-04-11 09:33:49 -07:00
Ronald S. Bultje
33d94a843f Remove copying of coefficients and predictor in i8x8 RD loop.
The resulting values are never used.

Change-Id: I688caf30da9aab87aa280cce913eda4f33172293
2013-04-10 17:39:03 -07:00
Ronald S. Bultje
8fb5be48a6 Make usage of sb_type independent of literal values.
Change-Id: I0d12f9ef9d960df0172a1377f8e5236eb6d90492
2013-04-10 17:38:57 -07:00
Ronald S. Bultje
b4f6098ef7 Make RD superblock mode search size-agnostic.
Merge various super_block_yrd and super_block_uvrd versions into one
common function that works for all sizes. Make transform size selection
size-agnostic also. This fixes a slight bug in the intra UV superblock
code where it used the wrong transform size for txsz > 8x8, and stores
the txsz selection for superblocks properly (instead of forgetting it).
Lastly, it removes the trellis search that was done for 16x16 intra
predictors, since trellis is relatively expensive and should thus only
be done after RD mode selection.

Gives basically identical results on derf (+0.009%).

Change-Id: If4485c6f0a0fe4038b3172f7a238477c35a6f8d3
2013-04-10 16:50:30 -07:00
Ronald S. Bultje
1932828d19 Merge "Make SB coding size-independent." into experimental 2013-04-10 08:51:58 -07:00
Ronald S. Bultje
a3874850dd Make SB coding size-independent.
Merge sb32x32 and sb64x64 functions; allow for rectangular sizes. Code
gives identical encoder results before and after. There are a few
macros for rectangular block sizes under the sbsegment experiment; this
experiment is not yet functional and should not yet be used.

Change-Id: I71f93b5d2a1596e99a6f01f29c3f0a456694d728
2013-04-09 21:28:27 -07:00
Jingning Han
12bf0796e6 Clamp inferred motion vectors only
Clamp only the motion vectors inferred from neighboring reference
macroblocks. The motion vectors obtained through motion search in
NEWMV mode are constrained during the search process, which allows
a relatively larger referencing region than the inferred mvs.
Hence further clamping the best mv provided by the motion search may
affect the efficacy of NEWMV mode.

Synchronized the decoding process. The decoded mvs in NEWMV modes
should be guaranteed to fit in the effective range. Put a mv range
clamping function there for security purpose.

This improves the coding performance of high motion sequences, e.g.,
derf set:
foreman 0.233%
husky   0.175%
icd     0.135%
mother_daughter 0.337%
pamphlet        0.561%

stdhd set:
blue_sky 0.408%
city     0.455%
also saw sunflower goes down by -0.469%.

Change-Id: I3fcbba669e56dab779857a8126a91b926e899cb5
2013-04-08 11:37:03 -07:00
John Koleszar
fa135d7b9e Merge changes Ibbfa68d6,Idb76a0e2 into experimental
* changes:
  Move EOB to per-plane data
  Move qcoeff, dqcoeff from BLOCKD to per-plane data
2013-04-05 15:56:50 -07:00
Ronald S. Bultje
36c3a67c20 Remove full-pixel-related code.
This is a VP8-only feature (part of profile 3) that is unsupported in
VP9.

Change-Id: I78016eede8d9c834d44d4c517f3e8b8fc2a378b1
2013-04-05 12:50:19 -07:00
John Koleszar
05a79f2fbf Move EOB to per-plane data
Continue migrating data from BLOCKD/MACROBLOCKD to the per-plane
structures.

Change-Id: Ibbfa68d6da438d32dcbe8df68245ee28b0a2fa2c
2013-04-04 21:30:23 -07:00
John Koleszar
4c05a051ab Move qcoeff, dqcoeff from BLOCKD to per-plane data
Start grouping data per-plane, as part of refactoring to support
additional planes, and chroma planes with other-than 4:2:0
subsampling.

Change-Id: Idb76a0e23ab239180c818025bae1f36f1608bb23
2013-04-04 16:30:57 -07:00
Deb Mukherjee
73031aaa7d Bugfix in encode_inter_mb_segment_8x8
Fixes an indexing bug. Looks like the bug has been there for a while.

Change-Id: I9fc04b0c30754bcb47366ad94a08112925600c4d
2013-04-04 11:07:19 -07:00
John Koleszar
a417a6e32c Merge "Removing redundant function arguments." into experimental 2013-04-01 21:09:48 -07:00
Deb Mukherjee
e3955007df Merge "Framework changes in nzc to allow more flexibility" into experimental 2013-03-29 15:57:27 -07:00
Deb Mukherjee
fe9b5143ba Framework changes in nzc to allow more flexibility
The patch adds the flexibility to use standard EOB based coding
on smaller block sizes and nzc based coding on larger blocksizes.
The tx-sizes that use nzc based coding and those that use EOB based
coding are controlled by a function get_nzc_used().
By default, this function uses nzc based coding for 16x16 and 32x32
transform blocks, which seem to bridge the performance gap
substantially.

All sets are now lower by 0.5% to 0.7%, as opposed to ~1.8% before.

Change-Id: I06abed3df57b52d241ea1f51b0d571c71e38fd0b
2013-03-28 09:33:50 -07:00
Ronald S. Bultje
9eea9fa206 Fix mix-up in pt token indexing.
This fixes uninitialized reads in the trellis, and probably makes the
trellis do something again.

Change-Id: Ifac8dae9aa77574bde0954a71d4571c5c556df3c
2013-03-28 09:24:29 -07:00
Dmitry Kovalev
17cddb4e26 Removing redundant function arguments.
Almost all arguments for vp9_build_inter32x32_predictors_sb and
vp9_build_inter64x64_predictors_sb can be deduced from the first macroblock
argument.

Change-Id: I5d477a607586d05698d5b3b9b9bc03891dd3fe83
2013-03-27 16:19:27 -07:00
Ronald S. Bultje
7c70145914 Merge "Add col/row-based coefficient scanning patterns for 1D 8x8/16x16 ADSTs." into experimental 2013-03-26 19:17:08 -07:00
Ronald S. Bultje
3c77ab4c0f Merge "Redo banding for all transforms." into experimental 2013-03-26 19:16:44 -07:00
Ronald S. Bultje
c6efbbcfe4 Merge "Use above/left (instead of previous in scan-order) as token context." into experimental 2013-03-26 19:16:24 -07:00
Deb Mukherjee
23144d2345 Implicit weighted prediction experiment
Adds an experiment to use a weighted prediction of two INTER
predictors, where the weight is one of (1/4, 3/4), (3/8, 5/8),
(1/2, 1/2), (5/8, 3/8) or (3/4, 1/4), and is chosen implicitly
based on consistency of the predictors to the already
reconstructed pixels to the top and left of the current macroblock
or superblock.

Currently the weighting is not applied to SPLITMV modes, which
default to the usual (1/2, 1/2) weighting. However the code is in
place controlled by a macro. The same weighting is used for Y and
UV components, where the weight is derived from analyzing the Y
component only.

Results (over compound inter-intra experiment)
derf: +0.18%
yt: +0.34%
hd: +0.49%
stdhd: +0.23%

The experiment suggests bigger benefit for explicitly signaled weights.

Change-Id: I5438539ff4485c5752874cd1eb078ff14bf5235a
2013-03-26 16:58:56 -07:00
Ronald S. Bultje
d9094d8fd3 Add col/row-based coefficient scanning patterns for 1D 8x8/16x16 ADSTs.
These are mostly just for experimental purposes. I saw small gains (in
the 0.1% range) when playing with this on derf.

Change-Id: Ib21eed477bbb46bddcd73b21c5c708a5b46abedc
2013-03-26 16:46:13 -07:00
Ronald S. Bultje
3120dbddb1 Redo banding for all transforms.
Now that the first AC coefficient in both directions use the same DC
as their context, there no longer is a purpose in letting both have
their own band. Merging these two bands allows us to split bands for
some of the very high-frequency AC bands.

In addition, I'm redoing the banding for the 1D-ADST col/row scans. I
don't think the old banding made any sense at all (it merged the last
coefficient of the first row/col in the same band as the first two of
the second row/col), which was clearly an oversight from the band being
applied in scan-order (rather than in their actual position). Now,
coefficients at the same position will be in the same band, regardless
what scan order is used. I think this makes most sense for the purpose
of banding, which is basically "predict energy for this coefficient
depending on the energy of context coefficients" (i.e. pt).

After full re-training, together with previous patch, derf gains about
1.2-1.3%, and hd/stdhd gain about 0.9-1.0%.

Change-Id: I7a0cc12ba724e88b278034113cb4adaaebf87e0c
2013-03-26 16:46:13 -07:00
Ronald S. Bultje
790fb13215 Use above/left (instead of previous in scan-order) as token context.
Pearson correlation for above or left is significantly higher than for
previous-in-scan-order (absolute values depend on position in scan, but
in general, we gain about 0.1-0.2 by using either above or left; using
both basically just makes this even better). For eob branch skipping,
we continue to use the previous token in scan order.

This helps about 0.9% on derf after re-training on a limited data set.
Full re-training and results on larger-resolution clips are pending.

Note that this commit breaks trellis, so we can probably get further
gains out of it by fixing trellis at some later point.

Change-Id: Iead68e296fc3a105cca746b5e3da9555d6010cfe
2013-03-26 16:46:09 -07:00
Dmitry Kovalev
56f3a2c663 Code cleanup: lower case variable names.
Renaming Width to width, Height to height and Version to version in
several structs and function signatures.

Change-Id: I084c3f7e747cb2ce3345aff27a3dff9b13a87543
2013-03-20 16:41:30 -07:00
Paul Wilkins
d8ffee4526 Changes to rd error_per_bit calculation.
Specifically changes to retain more precision
especially at low Q through to the point of use.

Change-Id: Ief5f010f2ca4daaabef49520e7edb46c35daf397
2013-03-18 23:07:51 +00:00
Paul Wilkins
ef179bce61 Merge "Adapt ARNR filter length and strength." into experimental 2013-03-18 12:00:39 -07:00
John Koleszar
c5b317057b Merge "Fix pulsing issue with scaling" into experimental 2013-03-18 11:57:36 -07:00
Paul Wilkins
cdb322dd72 Adapt ARNR filter length and strength.
Adjust the filter length and strength for each
ARF group based on a measure of difficulty (the boost)
and the active q range.

Remove lower limit on RDMULT value.

Average gains on the different sets in range 0.4%-0.9%.
However the ARNR changes give a very big boost on a
few clips.

Eg. Soccer ~5%, in derf set and Cyclist ~ 10% in the std-hd set

Change-Id: I2078d78798e27ad2bcc2b32d703ea37b67412ec4
2013-03-18 16:17:04 +00:00
Deb Mukherjee
b1921b2f08 Context-pred fix to not use top/left on edges
This fix resolves some of the mismatch issues being seen
recently. While this is the right thing to do when tiling
is used for this experiment, it is not the underlying cause
of the the mismatches.
Something else is causing writing outside of the allowable
frame area in the encoder leading to this mismatch.

Change-Id: If52c6f67555aa18ab8762865384e323b47237277
2013-03-16 09:26:52 -07:00
John Koleszar
9b7be88883 Fix pulsing issue with scaling
Updates the YV12_BUFFER_CONFIG structure to be crop-aware. The
exiting width/height parameters are left unchanged, storing the
width and height algined to a 16 byte boundary. The cropped
dimensions are added as new fields.

This fixes a nasty visual pulse when switching between scaled and
unscaled frame dimensions due to a mismatch between the scaling
ratio and the 16-byte aligned sizes.

Change-Id: Id4a3f6aea6b9b9ae38bdfa1b87b7eb2cfcdd57b6
2013-03-13 19:10:10 -07:00
Deb Mukherjee
a28139c849 Continued experiment with nonzero count
Adds probability updates for extra bits for the nzcs, code for
getting nzc stats, plus some minor cleanups and fixes.

Change-Id: If2814e7f04fb52f5025ad9f400f3e6c50a00b543
2013-03-08 16:37:08 -08:00
Ronald S. Bultje
0643c3f133 Merge "Add support for tx_select in i8x8 encoding in keyframes." into experimental 2013-03-08 16:25:27 -08:00
Jingning Han
2a5278bdbd Extend diff MV limit from +/-256 to +/-1024
Increase the motion search range by 4x. Change MV_CLASS tree of the
entropy coding to allow two additional mv classes to cover the
extended motion vector limit. The codec determines the effective
motion search range conditioned on the actual frame dimension.

It provides coding gains:

stdhd 0.39%
yt    0.56%
hd    0.47%

Major coding performance gains are packed in several sequences with
intense motion activities, e.g., ped_1080p gains 7% at high bit-rates,
and on average 3%.

TODO: Need to further tune the rate control and motion search units.

Change-Id: Ib842540a6796fbee5a797809433ef6a477c6d78d
2013-03-08 10:04:36 -08:00
Ronald S. Bultje
b41dee8428 Add support for tx_select in i8x8 encoding in keyframes.
Also enable tx_select for keyframes.

Change-Id: Iadb1231d9fa7af0c8dce3d9b41830b93a302479e
2013-03-08 09:28:46 -08:00
Ronald S. Bultje
d3724abe9f Re-add support for ADST in superblocks.
This also changes the RD search to take account of the correct block
index when searching (this is required for ADST positioning to work
correctly in combination with tx_select).

Change-Id: Ie50d05b3a024a64ecd0b376887aa38ac5f7b6af6
2013-03-07 11:19:10 -08:00
Deb Mukherjee
eb6ef2417f Coding con-zero count rather than EOB for coeffs
This patch revamps the entropy coding of coefficients to code first
a non-zero count per coded block and correspondingly remove the EOB
token from the token set.

STATUS:
Main encode/decode code achieving encode/decode sync - done.
Forward and backward probability updates to the nzcs - done.
Rd costing updates for nzcs - done.
Note: The dynamic progrmaming apporach used in trellis quantization
is not exactly compatible with nzcs. A suboptimal approach has been
used instead where branch costs are updated to account for changes
in the nzcs.

TODO:
Training the default probs/counts for nzcs

Change-Id: I951bc1e22f47885077a7453a09b0493daa77883d
2013-03-07 07:20:30 -08:00
Ronald S. Bultje
4209bba462 Merge changes Ifacbf5a0,Ibad7c3dd into experimental
* changes:
  vpxenc: actually report mismatch on stderr.
  Make superblocks independent of macroblock code and data.
2013-03-05 11:17:14 -08:00
Ronald S. Bultje
111ca42133 Make superblocks independent of macroblock code and data.
Split macroblock and superblock tokenization and detokenization
functions and coefficient-related data structs so that the bitstream
layout and related code of superblock coefficients looks less like it's
a hack to fit macroblocks in superblocks.

In addition, unify chroma transform size selection from luma transform
size (i.e. always use the same size, as long as it fits the predictor);
in practice, this means 32x32 and 64x64 superblocks using the 16x16 luma
transform will now use the 16x16 (instead of the 8x8) chroma transform,
and 64x64 superblocks using the 32x32 luma transform will now use the
32x32 (instead of the 16x16) chroma transform.

Lastly, add a trellis optimize function for 32x32 transform blocks.

HD gains about 0.3%, STDHD about 0.15% and derf about 0.1%. There's
a few negative points here and there that I might want to analyze
a little closer.

Change-Id: Ibad7c3ddfe1acfc52771dfc27c03e9783e054430
2013-03-04 16:34:36 -08:00
Dmitry Kovalev
49b697d327 Merge "Code cleanup." into experimental 2013-03-04 15:41:15 -08:00
Dmitry Kovalev
135428e954 Code cleanup.
Removing redundant 'extern' keyword, lowercase variable names.

Change-Id: I608e8d8579aba8981f5fac3493f77b4481b13808
2013-03-01 17:39:31 -08:00
John Koleszar
69c67c9531 Merge master branch into experimental
Picks up some build system changes, compiler warning fixes, etc.

Change-Id: I2712f99e653502818a101a72696ad54018152d4e
2013-03-01 11:06:05 -08:00
Ronald S. Bultje
90932399b4 Merge "Move eob from BLOCKD to MACROBLOCKD." into experimental 2013-02-27 11:39:16 -08:00
Ronald S. Bultje
e8c74e2b70 Move eob from BLOCKD to MACROBLOCKD.
Consistent with VP8.

Change-Id: I8c316ee49f072e15abbb033a80e9c36617891f07
2013-02-27 11:00:55 -08:00
John Koleszar
800ad0b886 Use ref_frame_map vice active_ref_idx on the encoder
This patch makes the encoder's use of ref_frame_map and active_ref_idx
consistent with the decoder. ref_frame_map[] maps a reference buffer
index to its actual location in the yv12_fb array, since many
references may share an underlying buffer. active_ref_idx[] mirrors
cpi->{lst,gld,alt}_fb_idx, holding the active references in each
slot.

This also fixes a bug in setup_buffer_inter() where the incorrect
reference was used to populate the scaling factors.

Change-Id: Id3728f6d77cffcd27c248903bf51f9c3e594287e
2013-02-27 08:22:40 -08:00
John Koleszar
77f88e97fa Combined motion compensation with scaled predictors
This patch extends the previous support for using references of a
different resolution in ZEROMV mode to all inter prediction modes.
Subpixel based best-mv scoring is disabled when the reference frame
differs in resolution from the current frame.

Change-Id: Id4dc3e5e6692de98d9857fd56bfad3ac57e944ac
2013-02-27 08:22:39 -08:00
John Koleszar
eb939f45b8 Spatial resamping of ZEROMV predictors
This patch allows coding frames using references of different
resolution, in ZEROMV mode. For compound prediction, either
reference may be scaled.

To test, I use the resize_test and enable WRITE_RECON_BUFFER
in vp9_onyxd_if.c. It's also useful to apply this patch to
test/i420_video_source.h:

  --- a/test/i420_video_source.h
  +++ b/test/i420_video_source.h
  @@ -93,6 +93,7 @@ class I420VideoSource : public VideoSource {

     virtual void FillFrame() {
       // Read a frame from input_file.
  +    if (frame_ != 3)
       if (fread(img_->img_data, raw_sz_, 1, input_file_) == 0) {
         limit_ = frame_;
       }

This forces the frame that the resolution changes on to be coded
with no motion, only scaling, and improves the quality of the
result.

Change-Id: I1ee75d19a437ff801192f767fd02a36bcbd1d496
2013-02-26 23:54:23 -08:00
Ronald S. Bultje
96d260515a Merge "Merge cnvcontext experiment." into experimental 2013-02-26 19:39:39 -08:00
Ronald S. Bultje
db54e6774f Merge "Minor cosmetics in rdopt." into experimental 2013-02-26 19:39:28 -08:00
John Koleszar
25686fc22d Merge "Refactor inter recon functions to support scaling" into experimental 2013-02-26 11:45:28 -08:00
Dmitry Kovalev
998bed1d2c Merge "Changing pitch value meaning for fht and iht transforms." into experimental 2013-02-26 10:44:15 -08:00
Ronald S. Bultje
b1641150b1 Merge cnvcontext experiment.
Change-Id: I35e64998b25694a3bb4a62164bba3c03c1db4bc7
2013-02-26 10:40:15 -08:00
Ronald S. Bultje
71539eae2a Minor cosmetics in rdopt.
Change-Id: I62497dcf2074b4bb4787bf660e727e5cf1bf3472
2013-02-26 10:40:11 -08:00
Ronald S. Bultje
c4ae97911a Merge "make cost_coeffs to use combined context" into experimental 2013-02-26 10:32:01 -08:00
John Koleszar
6a4f708c25 Refactor inter recon functions to support scaling
Ensure that all inter prediction goes through a common code path
that takes scaling into account. Removes a bunch of duplicate
1st/2nd predictor code. Also introduces a 16x8 mode for 8x8
MVs, similar to the 8x4 trick we were doing before. This has an
unexpected effect with EIGHTTAP_SMOOTH, so it's disabled in that
case for now.

Change-Id: Ia053e823a8bc616a988a0af30452e1e75a739cba
2013-02-26 10:03:29 -08:00
Dmitry Kovalev
9bf3f75168 Changing pitch value meaning for fht and iht transforms.
Pitch now means the number of elements, not the number of bytes.

Change-Id: Idb9f2f012e39b09d596a3cc1802305a80b7c13af
2013-02-25 18:19:55 -08:00
Yaowu Xu
ecb03e9a3f make cost_coeffs to use combined context
Change-Id: Ia15f4244595fab49bffda0c651a750a8a9481d28
2013-02-25 17:01:33 -08:00
Jingning Han
77a3becf92 clean up forward and inverse hybrid transform
Rebased.

Remove the old matrix multiplication transform computation. The 16x16
ADST/DCT can be switched on/off and evaluated by setting ACTIVE_HT16
300/0 in vp9/common/vp9_blockd.h.

Change-Id: Icab2dbd18538987e1dc4e88c45abfc4cfc6e133f
2013-02-25 09:16:12 -08:00
Ronald S. Bultje
0c9e2e9a1d Split coefficient token tables intra vs. inter.
Change-Id: I5416455f8f129ca0f450d00e48358d2012605072
2013-02-23 07:33:46 -08:00
Paul Wilkins
c17672a33d Further changes to coefficient contexts.
This patch alters the balance of context between the
coefficient bands (reflecting the position of coefficients
within a transform blocks) and the energy of the previous
token (or tokens) within a block.

In this case the number of coefficient bands is reduced
but more previous token energy bands are supported.

Some initial rebalancing of the default tables has been
by running multiple derf clips at multiple data rates using
the ENTOPY_STATS macro. Further balancing needs to be
done using larger image formatsd especially in regard to
the bigger transform sizes which are not as well represented
in encodings of smaller image formats.

Change-Id: If9736e95c391e711b04aef6393d26f60f36e1f8a
2013-02-23 07:29:09 -08:00
Jingning Han
c67a20994f Merge "Forward butterfly hybrid transform" into experimental 2013-02-22 09:20:26 -08:00
Paul Wilkins
b5f3cb6e37 Merge "Experimental removal of over quant code" into experimental 2013-02-22 08:44:40 -08:00
Paul Wilkins
dbf4942046 Experimental removal of over quant code
The over quant code was added in VP8 post
bitstream freeze to allow compression to lower
data rates

In VP9 the real qualtizer range has been greatly
extended anyway.

Change-Id: I5d384fa5e9a83ef75a3df34ee30627bd21901526
2013-02-22 14:00:51 +00:00
Jingning Han
babbd5d170 Forward butterfly hybrid transform
This patch includes 4x4, 8x8, and 16x16 forward butterfly ADST/DCT
hybrid transform. The kernel of 4x4 ADST is sin((2k+1)*(n+1)/(2N+1)).
The kernel of 8x8/16x16 ADST is of the form sin((2k+1)*(2n+1)/4N).

Change-Id: I8f1ab3843ce32eb287ab766f92e0611e1c5cb4c1
2013-02-21 18:24:28 -08:00
Deb Mukherjee
048f593703 Merge "Refactoring of switchable filter search for speed" into experimental 2013-02-21 09:23:50 -08:00
Deb Mukherjee
28b1db9278 Refactoring of switchable filter search for speed
Refactors the switchable filter search in the rd loop to
improve encode speed.

Uses a piecewise approximation to a closed form expression to estimate
rd cost for a Laplacian source with a given variance and quantization
step-size.

About 40% encode time reduction is achieved.

Results (on a feb 12 baseline) show a slight drop:

derf: -0.019%
yt: +0.010%
std-hd: -0.162%
hd: -0.050%

Change-Id: Ie861badf5bba1e3b1052e29a0ef1b7e256edbcd0
2013-02-20 18:34:42 -08:00
Ronald S. Bultje
aa84c16da2 Minor cosmetic cleanups.
Change-Id: I13d8ae754827368755575dd699a087b3b11f5b16
2013-02-15 17:21:16 -08:00
Ronald S. Bultje
ebfdaa0e0b Prevent filling transform size cache with uninitialized values.
The 32x32 value in case of splitmv was uninitialized. this leads to
all kind of erratic behaviour down the line. Also fill in dummy values
for superblocks in keyframes (the values are currently unused, but we
run into integer overflows anyway, which makes detecting bad cases
harder). Lastly, in case we did not find any RD value at all, don't
set tx_diff to INT_MIN, but instead set it to zero (since if we couldn't
find a mode, it's unlikely that any particular transform would have made
that worse or better; rather, it's likely equally bad for all tx_sizes).

Change-Id: If236fd3aa2037e5b398d03f3b1978fbbc5ce740e
2013-02-15 17:21:16 -08:00
Ronald S. Bultje
5bb103c486 Merge "Remove Y2 and Y-no-DC token types from the bitstream." into experimental 2013-02-15 17:11:20 -08:00
Jingning Han
e343732a92 Fixed a subtle issue that breaks encoding process
This issue breaks the encoding process of the codebase. The effect
emerges only in particular test sequence at certain bit-rates and
frame limits.

Change-Id: I02e080f2a49624eef9a21c424053dc2a1d902452
2013-02-15 14:49:30 -08:00
Ronald S. Bultje
3af36ea8cc Remove Y2 and Y-no-DC token types from the bitstream.
Change-Id: I7a5314daca993d46b8666ba1ec2ff3766c1e5042
2013-02-15 14:06:30 -08:00
Ronald S. Bultje
46dff5d233 Remove some Y2-related code.
Change-Id: I4f46d142c2a8d1e8a880cfac63702dcbfb999b78
2013-02-15 14:06:25 -08:00
Scott LaVarnway
ae886d6bff Moved vp9_get_coef_band to header file
allowing the compiler to inline.

Change-Id: I66e5caf5e7fefa68a223ff0603aa3f9e11e35dbb
2013-02-14 12:27:25 -08:00
Paul Wilkins
9255ad107f Abstract selection of coef band.
This patch abstracts the selection of the coefficient band
context into a function as a precursor to further experiments
with the coefficient context.

It also removes the large per TX size coefficient band structures
and uses a single matrix for all block sizes within the test function.

This may have an impact on quality (results to follow) but is only an
intermediate step in the process of redefining the context. Also the
quality impact will be larger initially because the default tables will
be out of step with the new banding.

In particular the 4x4 will in this case only use 7 bands. If needed we
can add back block size dependency localized within the function, but
this can follow on after the other changes to the definition of the
context.

Change-Id: Id7009c2f4f9bb1d02b861af85fd8223d4285bde5
2013-02-13 19:01:25 +00:00
Paul Wilkins
0d284ffed1 Abstract the selection of coefficient context.
This is an initial step to facilitate experimentation
with changes to the prior token context used to code
coefficients to take better account of the energy of
preceding tokens.

This patch merely abstracts the selection of context into
two functions and does not alter the output.

Change-Id: I117fff0b49c61da83aed641e36620442f86def86
2013-02-13 18:56:30 +00:00
Paul Wilkins
afa57bfc97 Merge "Remove NEWCOEFCONTEXT experiment." into experimental 2013-02-13 10:41:13 -08:00
Yaowu Xu
f01b08c96c Merge "enable bitstream lossless support" into experimental 2013-02-13 10:26:58 -08:00
Yaowu Xu
d3de97794f Merge "fix the lossless experiment" into experimental 2013-02-13 09:54:35 -08:00
Yaowu Xu
17db5d00be enable bitstream lossless support
1. Added a bit in frame header to  to indicate if a frame is encoded
in lossless mode, so decoder does not make the decision based on Q0
2. Minor changes to make sure that lossy coding works same as when
the lossless experiment is not enabled.
3. Renamed function pointers for transforms to be consistent, using
prefix fwd_txm and inv_txm for forward and inverse respectively

To encode in lossless mode, using "--lossless=1 --min-q=0 --max-q=0"
with vpxenc.

Change-Id: Ifae53b26d2ffbe378d707e29d96817b8a5e6c068
2013-02-13 09:24:39 -08:00
Yaowu Xu
16f25f9dc8 fix the lossless experiment
Change-Id: I95acfc1417634b52d344586ab97f0abaa9a4b256
2013-02-13 09:20:26 -08:00
Paul Wilkins
6a9f0c61a4 Remove NEWCOEFCONTEXT experiment.
Removal of the  NEWCOEFCONTEXT experiment to
reduce code clutter and make it easier to experiment with
some other changes to the coefficient coding context.

Change-Id: Icd17b421384c354df6117cc714747647c5eb7e98
2013-02-13 15:12:17 +00:00
Paul Wilkins
649be94cf0 Removal of Hybrid DWT/DCT experiment.
Removal of experiment to simplify code base for other
changes.

Change-Id: If0a33952504558511926ad212bc311fc2bffb19a
2013-02-13 15:08:48 +00:00
John Koleszar
1d60b6bcb5 Merge "Replace as_mv struct with array" into experimental 2013-02-12 13:59:04 -08:00
Jingning Han
57e995ff9c butterfly inverse 4x4 ADST
fixed format issues.

Implement the inverse 4x4 ADST using 9 multiplications. For this
particular dimension, the original ADST transform can be
factorized into simpler operations, hence is retained.

Change-Id: Ie5d9749942468df299ab74e90d92cd899569e960
2013-02-11 10:42:39 -08:00
John Koleszar
7ca517f755 Replace as_mv struct with array
Replace as_mv.{first, second} with a two element array, so that they
can easily be processed with an index variable.

Change-Id: I1e429155544d2a94a5b72a5b467c53d8b8728190
2013-02-08 20:23:35 -08:00
John Koleszar
dc836109e4 Merge "Pass macroblock index to pick inter functions" into experimental 2013-02-08 20:20:37 -08:00
Ronald S. Bultje
639b863d22 Make cost_coeffs() more efficient.
Cache the constant offset in one variable to prevent re-loading that
in each loop iteration, and mark the function as inline so we can use
the fact that the transform size is always known in the caller.

Almost 1% faster encoding overall.

Change-Id: Id78325a60b025057d8f4ecd9003a74086ccbf85a
2013-02-08 16:32:24 -08:00
John Koleszar
6125a1ed81 Pass macroblock index to pick inter functions
Pass the current mb row and column around rather than the
recon_yoffset and recon_uvoffset, since those offsets will
change from predictor to predictor, based on the reference
frame selection.

Change-Id: If3f9df059e00f5048ca729d3d083ff428e1859c1
2013-02-08 14:25:40 -08:00
John Koleszar
6dfc95fe63 Merge changes Icd1a2a5a,I204d17a1,I3ed92117 into experimental
* changes:
  Initial support for resolution changes on P-frames
  Avoid allocating memory when resizing frames
  Adds a test for the VP8E_SET_SCALEMODE control
2013-02-08 14:20:05 -08:00
John Koleszar
3de8ee6ba1 Merge changes Ife0d8147,I7d469716,Ic9a5615f into experimental
* changes:
  Restore SSSE3 subpixel filters in new convolve framework
  Convert subpixel filters to use convolve framework
  Add 8-tap generic convolver
2013-02-08 13:19:47 -08:00
John Koleszar
393b485627 Initial support for resolution changes on P-frames
Allows inter-frames to change resolution. Currently these are
almost equivalent to keyframes, as only intra prediction modes
are allowed, but without the other context resets that occur on
keyframes.

Change-Id: Icd1a2a5af0d9462cc792588427b0a1f5b12e40d3
2013-02-08 12:20:30 -08:00
Ronald S. Bultje
5cfd82bcaf Use fdct8x4 instead of fdct4x4 where the block size allows it.
This allows for faster SIMD implementations in the future (currently
there is no speed impact).

Change-Id: I732647e9148b5dcb44e6bc8728138f0141218329
2013-02-06 16:13:02 -08:00
Ronald S. Bultje
aac73df1a7 Use configure checks for various inline keywords.
Change-Id: I8508f1a3d3430f998bb9295f849e88e626a52a24
2013-02-06 16:12:56 -08:00
John Koleszar
31cbe2ed9a Eliminate tautology
Unreachable code
  that does nothing anyway
      removed forever.

Change-Id: I14105d2dd9dbc9d558f36464055e350dbeb45488
2013-02-06 08:22:59 -08:00
Paul Wilkins
8b4e9c5925 Merge "Change definition of NearestMV." into experimental 2013-02-06 04:06:31 -08:00
Ronald S. Bultje
1407bdc243 [WIP] Add column-based tiling.
This patch adds column-based tiling. The idea is to make each tile
independently decodable (after reading the common frame header) and
also independendly encodable (minus within-frame cost adjustments in
the RD loop) to speed-up hardware & software en/decoders if they used
multi-threading. Column-based tiling has the added advantage (over
other tiling methods) that it minimizes realtime use-case latency,
since all threads can start encoding data as soon as the first SB-row
worth of data is available to the encoder.

There is some test code that does random tile ordering in the decoder,
to confirm that each tile is indeed independently decodable from other
tiles in the same frame. At tile edges, all contexts assume default
values (i.e. 0, 0 motion vector, no coefficients, DC intra4x4 mode),
and motion vector search and ordering do not cross tiles in the same
frame.
t log

Tile independence is not maintained between frames ATM, i.e. tile 0 of
frame 1 is free to use motion vectors that point into any tile of frame
0. We support 1 (i.e. no tiling), 2 or 4 column-tiles.

The loopfilter crosses tile boundaries. I discussed this briefly with Aki
and he says that's OK. An in-loop loopfilter would need to do some sync
between tile threads, but that shouldn't be a big issue.

Resuls: with tiling disabled, we go up slightly because of improved edge
use in the intra4x4 prediction. With 2 tiles, we lose about ~1% on derf,
~0.35% on HD and ~0.55% on STD/HD. With 4 tiles, we lose another ~1.5%
on derf ~0.77% on HD and ~0.85% on STD/HD. Most of this loss is
concentrated in the low-bitrate end of clips, and most of it is because
of the loss of edges at tile boundaries and the resulting loss of intra
predictors.

TODO:
- more tiles (perhaps allow row-based tiling also, and max. 8 tiles)?
- maybe optionally (for EC purposes), motion vectors themselves
  should not cross tile edges, or we should emulate such borders as
  if they were off-frame, to limit error propagation to within one
  tile only. This doesn't have to be the default behaviour but could
  be an optional bitstream flag.

Change-Id: I5951c3a0742a767b20bc9fb5af685d9892c2c96f
2013-02-05 15:43:03 -08:00
John Koleszar
7a07eea13f Convert subpixel filters to use convolve framework
Update the code to call the new convolution functions to do subpixel
prediction rather than the existing functions. Remove the old C and
assembly code, since it is unused. This causes a 50% performance
reduction on the decoder, but that will be resolved when the asm for
the new functions is available.

There is no consensus for whether 6-tap or 2-tap predictors will be
supported in the final codec, so these filters are implemented in
terms of the 8-tap code, so that quality testing of these modes
can continue. Implementing the lower complexity algorithms is a
simple exercise, should it be necessary.

This code produces slightly better results in the EIGHTTAP_SMOOTH
case, since the filter is now applied in only one direction when
the subpel motion is only in one direction. Like the previous code,
the filtering is skipped entirely on full-pel MVs. This combination
seems to give the best quality gains, but this may be indicative of a
bug in the encoder's filter selection, since the encoder could
achieve the result of skipping the filtering on full-pel by selecting
one of the other filters. This should be revisited.

Quality gains on derf positive on almost all clips. The only clip
that seemed to be hurt at all datarates was football
(-0.115% PSNR average, -0.587% min). Overall averages 0.375% PSNR,
0.347% SSIM.

Change-Id: I7d469716091b1d89b4b08adde5863999319d69ff
2013-02-05 14:23:17 -08:00
Paul Wilkins
81043e8d62 Change definition of NearestMV.
This commit makes the NearestMV match the chosen
best reference MV. It can be a 0,0 or non zero vector
which means the the compound nearest mv mode can
combine a 0,0 and a non zero vector.

Change-Id: I2213d09996ae2916e53e6458d7d110350dcffd7a
2013-02-05 17:03:25 +00:00
Deb Mukherjee
a53be60904 Merge "Adding a frame parallel decoding mode" into experimental 2013-01-30 12:03:45 -08:00
Ronald S. Bultje
3febf9707d Default superblock skip flag to 32x32 for skip-blocks.
This is identical to the later decisions made in encode_superblock().
This commit doesn't actually change anything, but makes the mbmi state
more consistent between the RD loop and the final encode result.

Change-Id: I9e735afb7c5a52e5b61728cb88c67ef9b9bf59be
2013-01-29 21:46:31 -08:00
Ronald S. Bultje
b90996c51b Reset skip flag in superblock RD loop.
This is the superblock equivalent of commit 290b83a.

Change-Id: Ib3945dd9e992fa9ec1fdea5a11e17a3cc0e37637
2013-01-29 21:42:56 -08:00
Ronald S. Bultje
5a9da2d906 Merge "Fix block pointer corruption in intra8x8 prediction with 4x4 transform." into experimental 2013-01-29 12:49:42 -08:00
Paul Wilkins
5d1c62c639 Merge "Segment Skip Flag" into experimental 2013-01-29 09:29:26 -08:00
Ronald S. Bultje
ffc2e4f4af Fix block pointer corruption in intra8x8 prediction with 4x4 transform.
The RD loop would change the pointer after the first mode (DC) was tested,
leading to corrupt block objects being provided for the others. This
would essentially render the i8x8 predictor useless.

Change-Id: I16c5906ca64fb34878ac32ce59af8974e4582bb8
2013-01-29 09:18:47 -08:00
Paul Wilkins
0ff9b033b0 Segment Skip Flag
First step in simplifying the segment mode and
segment EOB flags into a simpler segment skip
flag that implies 0,0 mv and EOB at position 0.

Change-Id: Ib750cac31a7a02dc21082580498efd9f7d8d72a5
2013-01-28 17:28:04 +00:00
Deb Mukherjee
dfd89f2eab Adding a frame parallel decoding mode
Adds a flag to disable features that would inhibit frame parallel
decoding. This includes backward adaptation and MV sorting based
on search in ref frame buffer.

Also includes some minor clean-ups.

Change-Id: I434846717a47b7bcb244b37ea670c5cdf776f14d
2013-01-25 17:16:19 -08:00
Ronald S. Bultje
3ca5b35ce5 Merge "Remove "update_context" variable from VP9_COMP context." into experimental 2013-01-25 09:43:42 -08:00
Ronald S. Bultje
0a7b3953f0 Remove "update_context" variable from VP9_COMP context.
The variable is always zero.

Change-Id: Id5cdbecad543bca465a5b1d471badaec7e112c8d
2013-01-24 16:28:53 -08:00
Deb Mukherjee
01cafaab1d Adds an error-resilient mode with test
Adds an error-resilient mode where frames can be continued
to be decoded even when there are errors (due to network losses)
on a prior frame. Specifically, backward updates are turned off
and probabilities of various symbols are reset to defaults at
the beginning of each frame. Further, the last frame's mvs are
not used for the mv reference list, and the sorting of the
initial list based on search on previous frames is turned off
as well.

Also adds a test where an arbitrary set of frames are skipped
from decoding to simulate errors. The test verifies (1) that if
the error frames are droppable - i.e. frame buffer updates have
been turned off - there are no mismatch errors for the remaining
frames after the error frames; and (2) if the error-frames are non
droppable, there are not only no decoding errors but the mismatch
PSNR between the decoder's version of the post-error frames and the
encoder's version is at least 20 dB.

Change-Id: Ie6e2bcd436b1e8643270356d3a930e8989ff52a5
2013-01-23 21:56:15 -08:00
Frank Galligan
9ca907b53e libvpx: Fix some warnings.
Change-Id: If8be8b9d28a29631f29c46daea8a226ab3580610
2013-01-18 09:51:57 -08:00
John Koleszar
da832a80e4 Start to anonymize reference frames
Remove lst_fb_idx, gld_fb_idx, alt_fb_idx, refresh_last_frame,
refresh_golden_frame, refresh_alt_ref_frame from common. Gold/Alt are
encode side conventions. From the decoder's perspective, we want to be
dealing with numbered references.

Updates to active_ref 2 signal mode context switches, vestigial from
refresh_alt_ref_frame. This needs some clean up to make sense with
increased numbers of reference frames, as well as reimplementing the
swapping of alt/golden which was previously done using the
buffer-to-buffer copy mechanism removed in an earlier commit.

Change-Id: I7334445158b7666f9295d2a2dd22aa03f4485f58
2013-01-16 14:06:23 -08:00
Yaowu Xu
9bf73f46f9 fix a number issues that cause failures
During master jenkins verification proces

Change-Id: I3722b8753eaf39f99b45979ce407a8ea0bea0b89
2013-01-14 18:32:32 -08:00
John Koleszar
24bc1a7189 Use INT64_MAX instead of LLONG_MAX
These variables have the type int64_t, not long long. long long could
be a larger type than 64 bits. Emulate INT64_MAX for older versions of
MSVC, and remove the unreferenced vpx_ports/vpxtypes.h

Change-Id: Ideaca71838fcd3849d816d5ab17aa347c97d03b0
2013-01-14 15:57:21 -08:00
Ronald S. Bultje
c9071601a2 Remove compound intra-intra experiment.
This experiment gives little gains and adds relatively much code
complexity (and it hinders other experiments), so let's get rid of
it.

Change-Id: Id25e79a137a1b8a01138aa27a1fa0ba4a2df274a
2013-01-14 15:47:25 -08:00
Paul Wilkins
e2c696a7aa Merge "Fix compiler warnings" into experimental 2013-01-14 14:20:57 -08:00
Yaowu Xu
113005b11d Fix compiler warnings
The warnings caused verify failure with gerrit for several  commits

Change-Id: I030df8638bd69b8783a3ac58e720ff9f0bfd546c
2013-01-14 13:56:52 -08:00
Ronald S. Bultje
290b83ab62 Reset x->skip for each iteration in the RD loop.
This prevents ill-defined behaviour, such as setting x->skip for a mode
that is excluded because of frame-level flags (e.g. filter selection,
compound prediction selection), then not breaking out of the RD loop
because the mode is not allowed, but keeping the flag on. Whatever mode
is iterated through next in the RD loop will then carry this flag, and
all sort of bad stuff happens, such as x->skip being set on intra pred
modes.

Change-Id: I5bec46b36e38292174acb1c564b3caf00a9b4b9a
2013-01-14 12:44:32 -08:00
Paul Wilkins
d27ae620bc Remove INT64_MAX references.
Replace INT64_MAX references with LLONG_MAX
for windows build.

Change-Id: Ib8b45c1e9c15c043b2f54c27ed83b8682b2be34f
2013-01-11 19:45:26 +00:00
Ronald S. Bultje
aa2effa954 Merge tx32x32 experiment.
Change-Id: I615651e4c7b09e576a341ad425cf80c393637833
2013-01-10 08:23:59 -08:00
Ronald S. Bultje
6884a83f06 Merge superblocks64 experiment.
Change-Id: If6c88752dffdb566f8d4322f135145270716fb8e
2013-01-09 17:21:40 -08:00
Adrian Grange
7d6b5425d7 New prediction filter
This patch removes the old pred-filter experiment and replaces it
with one that is implemented using the switchable filter framework.

If the pred-filter experiment is enabled, three interopolation
filters are tested during mode selection; the standard 8-tap
interpolation filter, a sharp 8-tap filter and a (new) 8-tap
smoothing filter.

The 6-tap filter code has been preserved for now and if the
enable-6tap experiment is enabled (in addition to the pred-filter
experiment) the original 6-tap filter replaces the new 8-tap smooth
filter in the switchable mode.

The new experiment applies the prediction filter in cases of a
fractional-pel motion vector. Future patches will apply the filter
where the mv is pel-aligned and also to intra predicted blocks.

Change-Id: I08e8cba978f2bbf3019f8413f376b8e2cd85eba4
2013-01-09 12:00:39 -08:00
Deb Mukherjee
4b7304ee68 Adds 64x64 hybrid dct/dwt transform
This is to add to the 64x64 transform experiment as an alternative to
a 64x64 DCT.
Two levels of wavelet decomposition is used on a 64x64 block, followed
by 16x16 DCT on the four lowest subbands. The highest three subbands
are left untransformed after the first level DWT.

Change-Id: I3d48d5800468d655191933894df6b46e15adca56
2013-01-08 14:05:58 -08:00
Ronald S. Bultje
4455036cfc Merge superblocks (32x32) experiment.
Change-Id: I0df99742029834a85c4933652b0587cf5b6b2587
2013-01-08 12:54:45 -08:00
John Koleszar
879cb7d962 Merge vp9-preview changes into experimental branch
Incorportate vp9-preview changes by merging master branch into experimental.

Conflicts:
	test/test.mk
	vp9/common/vp9_filter.c
	vp9/common/vp9_idctllm.c
	vp9/common/vp9_invtrans.h
	vp9/common/vp9_mbpitch.c
	vp9/common/vp9_rtcd_defs.sh
	vp9/common/vp9_systemdependent.h
	vp9/common/vp9_type_aliases.h
	vp9/common/x86/vp9_asm_stubs.c
	vp9/common/x86/vp9_subpixel_mmx.asm
	vp9/decoder/vp9_decodframe.c
	vp9/decoder/vp9_dequantize.c
	vp9/decoder/vp9_dequantize.h
	vp9/decoder/vp9_onyxd_int.h
	vp9/encoder/vp9_bitstream.c
	vp9/encoder/vp9_encodeframe.c
	vp9/encoder/vp9_rdopt.c

Change-Id: I17f51c3666d1b59cf1a699f87607cbc5d30a87c5
2013-01-08 10:19:59 -08:00
Ronald S. Bultje
c13d9fef42 Re-enable support for static_threshold (encode_breakout).
Change-Id: Ibd7380f478d3127f9db91d0a4fd2fd0dfde961ab
2013-01-07 11:02:14 -08:00
Ronald S. Bultje
e6216d163a Don't use tx32x32 for macroblocks.
Change-Id: Ib674e0153ca360867ab7a20ba291ac9171a01250
2013-01-07 09:40:19 -08:00
Ronald S. Bultje
c3941665e9 64x64 blocksize support.
3.2% gains on std/hd, 1.0% gains on hd.

Change-Id: I481d5df23d8a4fc650a5bcba956554490b2bd200
2013-01-05 18:20:25 -08:00
Adrian Grange
81d1171fd4 Fix mode selection infinite loop bug
Mode selection for SBs could enter an infinite loop because
the interpolation filter mode index was not being reset
correctly.

Change-Id: I4bbe726f29ef5b6836e94884067c46084713cc11
2013-01-04 09:00:47 -08:00
Yaowu Xu
df7ce5a711 Merge "make cost_coeffs() and tokenize_b() consistent" into experimental 2013-01-03 09:57:07 -08:00
Yaowu Xu
818f5698fb Merge "Merge cost_coeffs_2x2() into cost_coeffs()" into experimental 2013-01-03 09:33:21 -08:00
Yaowu Xu
83664f457b make cost_coeffs() and tokenize_b() consistent
Change-Id: I7cdb5c32a1400f88ec36d08ea982e38b77731602
2013-01-03 09:31:47 -08:00
Adrian Grange
259b800832 New interpolation filter selection algorithm
Old Scheme:
When SWITCHABLE filter selection is enabled the encoder
evaluates the use of each interpolation filter type and
selects the best one to use at the MB level. A frame-
level flag can be set to force the use of a particular
filter type for all MBs in a frame if it is more efficient
to encode that way. The logic here involved a Q dependent
threshold that assumed that the second 8-tap filter was
a high-pass filter. However, this requires a trip around
the recode loop. If the frame-level flag indicates use
of a particular filter, the other filters are not
evaluated in the pick_mode loop.

New Scheme:
Each filter type is evaluated at the MB level and a record
of the best filter is kept, irrespective of what filter
is signaled at the frame-level. Once all MBs have been
encoded, a decision is made as to what frame-level mode
to set for the *next* frame. If one filter is used by 80%
or more of the MBs, then this filter is forced since it
is assumed that this will be more efficient if the
next frame has similar characteristics. i.e. there is a
one-frame lag between measuring the filter selection and
setting the frame-level mode to use.

Change-Id: I6a7e7ced8f27e120fafb99db2dc9c6293f8d20f7
2013-01-03 08:12:43 -08:00
Yaowu Xu
bd28510ef9 Merge cost_coeffs_2x2() into cost_coeffs()
Remove special case function cost_coeffs_2x2() and change function
cost_coeffs() to handle 2nd order haar block as it is handle all
other block types already.

Change-Id: I2aac6f81ee0ae9e03d6a8da4f8681d69b79ce41f
2013-01-03 08:00:00 -08:00
Paul Wilkins
cad4a91429 Change INT64_MAX to LLONG_MAX
This is needed to make the windows build work after
the removal of vp9_type_alisases.h.

Change-Id: I8addf38e9f3c8b864e0e30a8916a26e0264dd02c
2013-01-02 18:06:00 +00:00
Paul Wilkins
313d1100af Added update-able mv-ref probabilities.
Part of NEW_MVREF experiment.
Added update-able probabilities.

Change-Id: I5a4fcf4aaed1d0d1dac980f69d535639a3d59401
2013-01-02 14:22:11 +00:00
John Koleszar
5ebe94f9f1 Build fixes to merge vp9-preview into master
Various fixups to resolve issues when building vp9-preview under the more stringent
checks placed on the experimental branch.

Change-Id: I21749de83552e1e75c799003f849e6a0f1a35b07
2012-12-26 11:21:09 -08:00
Deb Mukherjee
08f0c7cc9c New previous coef context experiment
Adds an experiment to derive the previous context of a coefficient
not just from the previous coefficient in the scan order but from a
combination of several neighboring coefficients previously encountered
in scan order.  A precomputed table of neighbors for each location
for each scan type and block size is used. Currently 5 neighbors are
used.

Results are about 0.2% positive using a strategy where the max coef
magnitude from the 5 neigbors is used to derive the context.

Change-Id: Ie708b54d8e1898af742846ce2d1e2b0d89fd4ad5
2012-12-19 18:49:39 -08:00
Ronald S. Bultje
4cca47b538 Use standard integer types for pixel values and coefficients.
For coefficients, use int16_t (instead of short); for pixel values in
16-bit intermediates, use uint16_t (instead of unsigned short); for all
others, use uint8_t (instead of unsigned char).

Change-Id: I3619cd9abf106c3742eccc2e2f5e89a62774f7da
2012-12-18 15:31:19 -08:00
Ronald S. Bultje
5cab8b7a18 Merge "Give 4x4 scan and coef_band tables a _4x4 suffix." into experimental 2012-12-18 14:17:46 -08:00
Yunqing Wang
779c5f28a8 Fix uninitialized warning
Fixed uninitialized warning for txfm_size.

Change-Id: I42b7e802c3e84825d49f34e632361502641b7cbf
2012-12-18 13:19:04 -08:00
Ronald S. Bultje
8986eb5c26 Give 4x4 scan and coef_band tables a _4x4 suffix.
This matches the names of tables for all other transform sizes.

Change-Id: Ia7681b7f8d34c97c27b0eb0e34d490cd0f8d02c6
2012-12-18 10:49:10 -08:00
Paul Wilkins
d8f5d1b257 Problem of over smoothing with intra modes.
In some cases intra modes in inter frames give
an over smoothed appearance. Especially with
noisy but flat content.

Also in some cases there were problems with key
frame sizing again with very flat but noisy content.

These are temporary changes to help alleviate the
visual problems but will almost certainly hurt metric
results especially at the very low data rate end.

Change-Id: I11549179a19277ffc283d9788bc70168f2a8bdc9
2012-12-17 11:54:17 +00:00
Deb Mukherjee
7fa3deb1f5 Build fixes with teh super blcoks and 32x32 expts
Change-Id: I3c751f8d57ac7d3b754476dc6ce144d162534e6d
2012-12-13 12:18:38 -08:00
Scott LaVarnway
b575394e21 Improved vp9_ihtllm_c
As suggested by Yaowu, we can use eob to reduce the complexity
of the vp9_ihtllm_c function.  For the 1080p test clip used, the decoder
performance improved by 17%.

Change-Id: I32486f2f06f9b8f60467d2a574209aa3a3daa435
2012-12-12 15:49:39 -08:00
Ronald S. Bultje
39de1e14ed Merge "Consistently use get_prob(), clip_prob() and newly added clip_pixel()." into experimental 2012-12-12 10:34:14 -08:00
Ronald S. Bultje
4d0ec7aacd Consistently use get_prob(), clip_prob() and newly added clip_pixel().
Add a function clip_pixel() to clip a pixel value to the [0,255] range
of allowed values, and use this where-ever appropriate (e.g. prediction,
reconstruction). Likewise, consistently use the recently added function
clip_prob(), which calculates a binary probability in the [1,255] range.
If possible, try to use get_prob() or its sister get_binary_prob() to
calculate binary probabilities, for consistency.

Since in some places, this means that binary probability calculations
are changed (we use {255,256}*count0/(total) in a range of places,
and all of these are now changed to use 256*count0+(total>>1)/total),
this changes the encoding result, so this patch warrants some extensive
testing.

Change-Id: Ibeeff8d886496839b8e0c0ace9ccc552351f7628
2012-12-12 10:01:19 -08:00
Yaowu Xu
0c35b27689 Merge "clean up tokenize_b() and stuff_b()" into experimental 2012-12-11 13:51:56 -08:00
Yaowu Xu
899f0fc126 clean up tokenize_b() and stuff_b()
Change-Id: I0c1be01aae933243311ad321b6c456adaec1a0f5
2012-12-11 13:32:16 -08:00
Paul Wilkins
d124465975 Further changes to mv reference code.
Some further changes and refactoring of mv
reference code and selection of center point for
searches. Mainly relates to not passing so many
different local copies of things around.

Some place holder comments.

Change-Id: I309f10ffe9a9cde7663e7eae19eb594371c8d055
2012-12-10 17:31:51 +00:00
Ronald S. Bultje
885cf816eb Introduce vp9_coeff_probs/counts/stats/accum types.
Use these, instead of the 4/5-dimensional arrays, to hold statistics,
counts, accumulations and probabilities for coefficient tokens. This
commit also re-allows ENTROPY_STATS to compile.

Change-Id: If441ffac936f52a3af91d8f2922ea8a0ceabdaa5
2012-12-07 16:09:59 -08:00
Ronald S. Bultje
c456b35fdf 32x32 transform for superblocks.
This adds Debargha's DCT/DWT hybrid and a regular 32x32 DCT, and adds
code all over the place to wrap that in the bitstream/encoder/decoder/RD.

Some implementation notes (these probably need careful review):
- token range is extended by 1 bit, since the value range out of this
  transform is [-16384,16383].
- the coefficients coming out of the FDCT are manually scaled back by
  1 bit, or else they won't fit in int16_t (they are 17 bits). Because
  of this, the RD error scoring does not right-shift the MSE score by
  two (unlike for 4x4/8x8/16x16).
- to compensate for this loss in precision, the quantizer is halved
  also. This is currently a little hacky.
- FDCT and IDCT is double-only right now. Needs a fixed-point impl.
- There are no default probabilities for the 32x32 transform yet; I'm
  simply using the 16x16 luma ones. A future commit will add newly
  generated probabilities for all transforms.
- No ADST version. I don't think we'll add one for this level; if an
  ADST is desired, transform-size selection can scale back to 16x16
  or lower, and use an ADST at that level.

Additional notes specific to Debargha's DWT/DCT hybrid:
- coefficient scale is different for the top/left 16x16 (DCT-over-DWT)
  block than for the rest (DWT pixel differences) of the block. Therefore,
  RD error scoring isn't easily scalable between coefficient and pixel
  domain. Thus, unfortunately, we need to compute the RD distortion in
  the pixel domain until we figure out how to scale these appropriately.

Change-Id: I00386f20f35d7fabb19aba94c8162f8aee64ef2b
2012-12-07 14:45:05 -08:00
Deb Mukherjee
8b92f1e023 Supports inter-intra prediction with superblocks
Adds support for compound inter-intra prediction with superblocks.
Also, fixes a bug that disabled intra modes for superblocks.

Change-Id: I4d711317e1bc19df8c2f32dc645429f7fff31036
2012-12-01 15:19:55 -08:00
Deb Mukherjee
6632330702 Adds switchable filters with superblocks
Allows switchbale filters to be used without mismatch when the
superblock experiment is on.

Also removes a spurious clamping code in decodemv.c which causes
rare encode/decode mismatches.

Change-Id: I809d9ee0b2859552b613500b539a615515b863ae
2012-11-30 09:37:08 -08:00
Jim Bankoski
9f9370425b warnings in various experiments
Change-Id: Ib5106d4772450f8026f823dd743f162ab833b1d6
2012-11-30 07:31:37 -08:00
Yaowu Xu
ff2f9de828 Merge changes Iaa67bcf1,Ibea3bc80 into experimental
* changes:
  more warning cleanup
  unused variables & warnings
2012-11-29 09:34:10 -08:00
Yaowu Xu
6431007df3 Merge "minor fix to eob check for setting CONTEXT" into experimental 2012-11-29 09:27:00 -08:00
Yaowu Xu
7ab1d3e49f minor fix to eob check for setting CONTEXT
Previously, the "!=" check is logically incorrect when eob is at 0 and
effective coefficient starting position is 1. This commit should have
no effect on bitstream.

Change-Id: I6ce3a847c7e72bfbe4f7c74f88e3310c6b9b6d30
2012-11-29 09:10:15 -08:00
Jim Bankoski
a802f5e783 unused variables & warnings
Change-Id: Ibea3bc80eb26a975faaa60268bbc93237f82bc57
2012-11-29 09:02:47 -08:00
Jim Bankoski
030e268a90 ihtllm moves to rtcd
clears up some warnings

Change-Id: I9899637497c6ad7519f098e055ab98580ae6d688
2012-11-29 07:19:38 -08:00
Jim Bankoski
13dbf1fb17 more rtcd cleanup
Change-Id: Ieefd76e164ca4aa87597da0412977614ddfbacb7
2012-11-28 17:27:15 -08:00
Deb Mukherjee
0742b1e4ae Fixing 8x8/4x4 ADST for intra modes with tx select
This patch allows use of 8x8 and 4x4 ADST correctly for Intra
16x16 modes and Intra 8x8 modes when the block size selected
is smaller than the prediction mode. Also includes some cleanups
and refactoring.

Rebase.

Change-Id: Ie3257bdf07bdb9c6e9476915e3a80183c8fa005a
2012-11-28 16:21:12 -08:00
Jim Bankoski
c67873989f fixed includes to be fully specified
Change-Id: Ia1cce221f8511561b9cbd8edb7726fbc286ff243
2012-11-28 10:53:17 -08:00
John Koleszar
fcccbcbb39 Add vp9_ prefix to all vp9 files
Support for gyp which doesn't support multiple objects in the same
static library having the same basename.

Change-Id: Ib947eefbaf68f8b177a796d23f875ccdfa6bc9dc
2012-11-27 14:12:30 -08:00