Commit Graph

147 Commits

Author SHA1 Message Date
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
bed27a960a Add a feature to reduce chrome intra mode search
Change-Id: I721ebdeef2b53ce3e5c3eba3f7462ae2103c95a8
2013-07-10 08:59:18 -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
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
Yaowu Xu
0d7b7c09cb Added a speed feature use_square_partition_only
This commit adds a speed feature where only squared partition are
evaluated in partition picking. Enable this feature in cpu-used 2
reduces encoding time by ~30%.

loss of compression:
-0.9% on cif set
-1.23% on stdhd

Change-Id: Ia6fad11210f0b78365abb889f9245604513be5b9
2013-07-02 16:40:15 -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
Yunqing Wang
b12e060b55 Add speed feature to disable splitmv
Added a speed feature in speed 1 to disable splitmv for HD (>=720)
clips. Test result on stdhd set: 0.3% psnr loss and 0.07% ssim
loss. Encoding speedup is 36%.

(For reference: The test result on derf set showed 2% psnr loss
and 1.6% ssim loss. Encoding speedup is 34%. SPLITMV should be
enabled for small resolution videos.)

Change-Id: I54f72b94f506c6d404b47c42e71acaa5374d6ee6
2013-07-02 10:02:34 -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
Jim Bankoski
d4158283e7 use partitioning from last frame
This cl converts use partition from last frame to do the following:

if part is none,horz, vert -> try split
if part != none and one of the children is not split - try none


Change-Id: I5b6c659e35f3ac9f11c051b92ba98af6d7e8aa87
Signed-off-by: Jim Bankoski <jimbankoski@google.com>
2013-07-01 18:18:50 -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
Dmitry Kovalev
59070f6e3c Merge "Removing CONFIG_DEBUG checks on assertions." 2013-06-28 14:03:28 -07:00
Dmitry Kovalev
8e6ce6bb9e Removing CONFIG_DEBUG checks on assertions.
Adding CHECK_MEM_ERROR macro to vp9_common.h and removing two duplicated
ones from vp9_onyx_int.h and vp9_onyxd_int.h.

Change-Id: I916afec61b3019f18193135dac7c35ed0f89b8b6
2013-06-28 10:36:20 -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
1374a06bd8 Optimize partition search order
This commit change the partition search order to allow checking of
rectangular partition to be done after square partitions. It also
added a speed feature to skip rectangular partition check when
NONE is better than SPLIT in RD sense.

This feature roughly speed up encoder by 1.5X with loss on compression
-0.91% on cif set
-0.56% on stdhd set

Change-Id: I0d2d06993041aa9ea9073fcc39c54f73a127dfa4
2013-06-28 07:13:54 -07:00
Paul Wilkins
05ffdf2625 Merge "Auto adapt step size feature." 2013-06-27 02:28:41 -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
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
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
Jim Bankoski
9f2a1ae23e adds force partitioning greater than or less than block size
adds a new speed feature to force partitioning to be greater than
or less than a certain size

Change-Id: I8c048eeeef93700ae822eccf98f8751a45b2e7d0
2013-06-20 09:51:42 -07:00
Jim Bankoski
18bdf708e7 adds a set partitioning to speed features
this feature lets you set a partitioning size to be used by the entire
frame.

Change-Id: I208a4c8c701375cbb054418266f677768b6f8f06
2013-06-20 09:50:44 -07:00
Jim Bankoski
476d73d294 partition by variance using var from last frame
This uses variance to split partition. Variance is calculated using
nearest mv,  always from last ref frame.

Change-Id: Idd015b4a9aa3bc82591759eac239680c07496896
2013-06-20 09:48:22 -07:00
Jim Bankoski
1f94b97694 convert all speed things to speed features
Change-Id: Ie24489a4d39f3e53e816eeebf75a1c9c7d94515a
2013-06-20 09:42:44 -07:00
Jim Bankoski
0fad6a9d99 fix to set up new speed feature
This uses the speed feature functionality for code.

Change-Id: I9cd16c0c5f98520ae27ebba81aa2c178546587f8
2013-06-20 09:35:02 -07:00
Ronald S. Bultje
8a0808a145 Fix row tiling.
Change-Id: I57be4eeaea6e4402f6a0cc04f5c6b7a5d9aedf9b
2013-06-12 13:42:59 -04:00
Deb Mukherjee
17da2cab78 TX_SIZE contexts simplification.
Reduces TX_SIZE contexts to 2 for each kind. The code is
cleaner and there is hardly any performance difference with
more than two contexts.

Results: almost neutral

Change-Id: I17656bd6db76224ae2856adf882504560e7dbaa4
2013-06-08 12:32:26 -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
Deb Mukherjee
869a39ba60 Cleans up mbskip encoding
Refactors mbskip coding to be compatible with coding of the rest of
the symbols. Adds forward/backward adaptation and removes a lot of
the legacy code.

Results:
fast50: +1.6%
derfraw300: +0.317%

Change-Id: I395a2976d15af044d3b8ded5acfa45f6f065f980
2013-06-07 16:00:26 -07:00
Yaowu Xu
0bb6da3668 Merge "Remove two un-used entries in mode_lf_delta[]" into experimental 2013-06-07 10:10:45 -07:00
Yaowu Xu
b097a3ba82 Remove two un-used entries in mode_lf_delta[]
With the removal of i4X4 and SPLIT_MV modes, the two entries for the
modes are no longer used. This patch remove the coding of the deltas.

Change-Id: Iea4eb500404ebe9706159380a03b8eca542fb4c3
2013-06-07 09:24:09 -07:00
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
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
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
Deb Mukherjee
83885235a7 Clean-ups on switchable interpolation and mv_ref
Adds backward adaptation and differential forward updates of switchable
interpolation filter probabilities. Also adds some cosmetic cleanups
and minor fixes on mv_ref probabilities.

derfraw300: +0.353% (with most coming from switchable interp changes)

Change-Id: Ie2718be73528c945fd0d80cfd63ca2d9cb3032de
2013-06-05 10:11:52 -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
Ronald S. Bultje
b480d413e7 Minor cosmetic changes.
Change-Id: Ieb4a8c97bf1b1dfb993f40a9a3ef3bed5ae7d948
2013-05-30 20:58:53 -07:00
Ronald S. Bultje
310bc1030a Merge "Merge VP9_YMODES, VP9_UV_MODES, INTRA_MODE_COUNT and cousins." into experimental 2013-05-30 20:58:19 -07:00
Ronald S. Bultje
7d549870f7 Merge "Remove TX_SIZE_MAX_MB." into experimental 2013-05-30 20:58:16 -07:00
Ronald S. Bultje
6ea6f4d253 Merge "Remove one (unused) entry from mvref tables." into experimental 2013-05-30 20:58:13 -07:00
Ronald S. Bultje
a433abbcad Merge VP9_YMODES, VP9_UV_MODES, INTRA_MODE_COUNT and cousins.
These are now merged in a new define called VP9_INTRA_MODES.

Change-Id: I0890f895756a7395d84c92f98f43e43f4cf9050d
2013-05-30 17:21:06 -07:00
Ronald S. Bultje
4d3d00b195 Remove TX_SIZE_MAX_MB.
Change-Id: I715870513d1fef8471bfd0f5218a79360a1ef126
2013-05-30 17:21:06 -07:00
Ronald S. Bultje
580d29bdbb Remove one (unused) entry from mvref tables.
Change-Id: Ieb4669ae564bec9f3051485ecdf186cb4e00decb
2013-05-30 17:21:06 -07:00
Deb Mukherjee
0807e1b8f6 Merge "Bugfix in forward update with modeling on." into experimental 2013-05-30 13:33:51 -07:00