Commit Graph

41 Commits

Author SHA1 Message Date
Yunqing Wang
38f1fbbb75 Migrate quantization functions from vp9/ to vpx_dsp/
The following quantization functions were moved:
vp9_quantize_b
vp9_quantize_b_32x32
vp9_highbd_quantize_b
vp9_highbd_quantize_b_32x32

vp9_quantize_dc
vp9_quantize_dc_32x32
vp9_highbd_quantize_dc
vp9_highbd_quantize_dc_32x32

The purpose of doing that was to allow these functions to be shared
by multiple codecs.

Change-Id: Id8ab939f283353cdd07bd930d47db3d932a5d87f
2015-07-17 16:38:14 -07:00
Deb Mukherjee
6910e92d04 dc quantizer fix for 32x32 transforms
The rounding factor needs to be scaled down by a factor of 2.
Also, the quantized and dequantized coefficients are memset to 0
when dc quantizer is used.

Change-Id: Ifa68bab02addbf1b83d249c5b4cbd5cda796b1cf
2015-03-03 15:58:27 -08:00
Jingning Han
d0f2377027 Revert "Revert "Removal of legacy zbin_extra / zbin_oq_value.""
This reverts commit 9946ee23e0.

Fix the ssse3 asm function.

Change-Id: I07f77a63aa98087626e45c4e87aa5dcafc0b0b07
2014-12-22 10:09:25 -08:00
Paul Wilkins
9946ee23e0 Revert "Removal of legacy zbin_extra / zbin_oq_value."
This reverts commit e9b586e21b.

Change-Id: I5b36e6727da6c05278d97e2c37b80c109f79bed4
2014-12-19 15:02:58 +00:00
Paul Wilkins
e9b586e21b Removal of legacy zbin_extra / zbin_oq_value.
zbin extra / zbin_oq_value was widely passed around,
hence removal touches a lot of code.

Change-Id: Idc94359735b60c38a160e4385ae09d5ca8b6b8e5
2014-12-18 16:49:11 +00:00
Paul Wilkins
60e9b731cf Remove mode dependent zbin boost.
Initial patch to remove get_zbin_mode_boost() and
cpi->zbin_mode_boost.

For now sets a dummy value of 0 for zbin extra pending
a further clean up patch.

Change-Id: I64a1e1eca2d39baa8ffb0871b515a0be05c9a6af
2014-12-18 16:45:52 +00:00
Deb Mukherjee
1929c9b391 Rename highbitdepth functions to use highbd prefix
Uses highbd_ prefix convention consistently.

Change-Id: I58f7f799a7ff8e32701bcd71c955bcf1cdd4581e
2014-10-09 14:40:40 -07:00
Deb Mukherjee
10783d4f3a Adds high bitdepth transform functions and tests
Adds various high bitdepth transform functions and tests.
Much of the changes are related to using typedefs tran_low_t
and tran_high_t for the final transform cofficients and intermediate
stages of the transform computation respectively rather than fixed
types int16_t/int. When vp9_highbitdepth configure flag is off,
these map tp int16_t/int32_t, but when the flag is on, they map
to int32_t/int64_t to make space for needed extra precision.

Change-Id: I3c56de79e15b904d6f655b62ffae170729befdd8
2014-09-11 19:56:33 -07:00
Alex Converse
5926e7c0e8 Remove unfinished VP9 alpha channel.
Change-Id: Ic5d3a3a0dac10b49495771886a31e793bb78b5ca
2014-07-21 15:55:50 -07:00
Jingning Han
9ac2f66320 Re-design quantization process
This commit re-designs the quantization process for transform
coefficient blocks of size 4x4 to 16x16. It improves compression
performance for speed 7 by 3.85%. The SSSE3 version for the
new quantization process is included.

The average runtime of the 8x8 block quantization is reduced
from 285 cycles -> 255 cycles, i.e., over 10% faster.

Change-Id: I61278aa02efc70599b962d3314671db5b0446a50
2014-07-01 17:00:07 -07:00
Jingning Han
ccba289f8d Fast computation path for forward transform and quantization
This commit enables a fast path computational flow for forward
transformation. It checks the sse and variance of prediction
residuals and decides if the quantized coefficients are all
zero, dc only, or more. It then selects the corresponding coding
path in the forward transformation and quantization stage.

It is currently enabled in rtc coding mode. Will do it for rd
coding mode next.

In speed -6, the runtime for pedestrian_area 1080p at 1000 kbps
goes down from 14234 ms to 13704 ms, i.e., about 4% speed-up.
Overall coding performance for rtc set is changed by -0.18%.

Change-Id: I0452da1786d59bc8bcbe0a35fdae9f623d1d44e1
2014-06-12 11:10:54 -07:00
Alex Converse
e6222b1a47 Fix the CONFIG_ALPHA build.
Change-Id: Ib89fe34812c17cd6294ce3c38f87d43a79abb16f
2014-04-22 11:23:49 -07:00
Dmitry Kovalev
d1a396d8b9 Moving q_trans[] table to vp9_quantize.{c, h}.
Change-Id: I1324c339815a47004ddccdaf651d24c60382b92f
2014-04-09 13:35:39 -07:00
Dmitry Kovalev
7bcfa31750 Moving encoder quantization parameters into separate struct.
Change-Id: I2a169535489aeda3943fb5a46ab53e7a12abaa36
2014-03-28 16:46:41 -07:00
Dmitry Kovalev
f527c46f20 Cleaning up vp9_quantize.c.
Change-Id: I9a38af32f16f196b83dd69755eafb9543edf5691
2014-02-28 10:11:31 -08:00
James Zern
b453941caf vp9/encoder: add extern "C" to headers
Change-Id: I4f51ce859a97bf1b8fd2b37ac585b7c643232b69
2014-01-23 16:21:24 -08:00
Alex Converse
2360a5f093 Remove plane_block_idx.
Its last remaining caller can be passed its results directly without any
additional work. Also, it's not non-4:2:0 safe.

Change-Id: Ia5089ba5f7f66c7617270483c619c9271aefd868
2013-12-02 18:33:50 -08:00
Dmitry Kovalev
8253532c2d Cleaning up vp9_regular_quantize_b_4x4.
Passing scan & iscan as parameters, adding useful local variables.

Change-Id: Ia2a87906941db9557350d273669ce5c3cdb7235d
2013-10-28 14:28:28 -07:00
Dmitry Kovalev
ec414372e8 Removing quantize_b_4x4 function pointer.
The pointer was asigned only once with vp9_regular_quantize_b_4x4, calling
this function directly now. Also removing unused declarations:
  prototype_quantize_block
  prototype_quantize_block_pair
  prototype_quantize_mb
  vp9_regular_quantize_b_4x4_pair
  vp9_regular_quantize_b_8x8

Change-Id: I14325bc2f082336820671eafbc06126651b79f73
2013-10-22 13:09:36 -07:00
Dmitry Kovalev
12d57a9409 Removing redundant 'extern' keyword.
Change-Id: Ie51306689c0dc527a8aa12d3984389dd8f360dea
2013-09-24 15:13:09 -07:00
Ronald S. Bultje
1ff94fea56 Inline vp9_quantize() in xform_quant().
Cycle times:
4x4:    151 to  131 cycles (15% faster)
8x8:    334 to  306 cycles (9% faster)
16x16: 1401 to 1368 cycles (2.5% faster)
32x32: 7403 to 7367 cycles (0.5% faster)

Total encode time of first 50 frames of bus @ 1500kbps (speed 0)
goes from 1min39.2 to 1min38.6, i.e. a 0.67% overall speedup.

Change-Id: I799a49460e5e3fcab01725564dd49c629bfe935f
2013-07-15 17:30:57 -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
John Koleszar
3f4e80634b Create common vp9_encode_sb{,y}
Creates a common encode (subtract, transform, quantize, optimize,
inverse transform, reconstruct) function for all sb sizes, including
the old 16x16 path.

Change-Id: I964dff1ea7a0a5c378046a069ad83495f54df007
2013-05-02 14:02:03 -07:00
Johann
e43662e8e6 Remove unused quantize optimizations.
Files were copied from vp8 and never maintained.

Change-Id: I9659a8755985da73e8c19c3c984423b6666d8871
2013-04-30 18:42:05 -07:00
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
13e41ba440 Remove unused macroblock versions of reconstruction functions.
More specifically, remove vp9_quantize_mb*, vp9_optimize_mb*,
vp9_inverse_transform_mb* and vp9_transform_mb*. Instead, use the
generic _sb* functions that take a size argument, and call them with
BLOCK_SIZE_MB16X16.

Change-Id: I33024afea95d3a23ffbc1df7da426e4645110f29
2013-04-11 12:27:15 -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
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
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
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
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
Ronald S. Bultje
e8c74e2b70 Move eob from BLOCKD to MACROBLOCKD.
Consistent with VP8.

Change-Id: I8c316ee49f072e15abbb033a80e9c36617891f07
2013-02-27 11:00:55 -08:00
Ronald S. Bultje
46dff5d233 Remove some Y2-related code.
Change-Id: I4f46d142c2a8d1e8a880cfac63702dcbfb999b78
2013-02-15 14:06:25 -08:00
Ronald S. Bultje
aa2effa954 Merge tx32x32 experiment.
Change-Id: I615651e4c7b09e576a341ad425cf80c393637833
2013-01-10 08:23:59 -08:00
Ronald S. Bultje
4455036cfc Merge superblocks (32x32) experiment.
Change-Id: I0df99742029834a85c4933652b0587cf5b6b2587
2013-01-08 12:54:45 -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
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
Johann
34591b54dd Remove ARM optimizations from VP9
Change-Id: I9f0ae635fb9a95c4aa1529c177ccb07e2b76970b
2012-12-03 12:50:15 -08:00
Jim Bankoski
2b8dc065d1 google style guide include guards
Change-Id: I2c252f3ddcc99e96c1f5d3dab8bcb25a2a3637ea
2012-11-30 07:30:59 -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