GitLab

intra-interp experiment allows intra prediction to use different
interpolation filters. It was part of the ext-intra in the nextgenv2
branch.

Change-Id: I27ab692494dc79bb92e457dbf9a72988577f1c6d

Make the reference motion vector search range and ranking weight
scalable with the mode_info size. It retains the same coding
statistics.

Change-Id: I238e0ff6032fea842c3b6094287398f7ef7cb261

This reverts commit 3e882f29.

This causes build and link issues under visual studio in both x86 and
x64 configs

Change-Id: Id8b1e7e2f313817f034afddc2c627db8aa741d14

- For all blocks with width >= 16.
- Add test_count to make the unit tests harder to pass.
- Speed testing on 1080p, 100 frames, 5 Mbps, CPU, i7-6700
  User level time reduction:
   baseline:                  3.68%
   baseline + ext-partition: 36.12%

Change-Id: I78c5d9ca216f0fd91f1a360dca2190b11fd54a08

Change-Id: I846cd0c4f5e3d60c76b969183d17ce11e6ba2999

This means that all shared PVQ code is now in fixed-point.
Applies the following patches from daala:

963d3a9 pvq: fix artificial chroma boost with --enable-float-pvq
7522712 pvq: cosmetics: drop OD_ROUND32 where it is a noop
3395f61 pvq: in float pvq case, mag must default to 1.0
29193bb pvq: fix typo
c774e25 pvq: convert od_init_qm to int
b5607d6 pvq: use small LUT for integer sqrt((n+3)/2) sqrt((n+2)/2)
a989c43 pvq: use od_beta_rcp in od_gain_compand
12c31ea pvq: use od_rcp_beta in od_compute_k

Change-Id: I0ad6a896c9f9fdd752f3b65a0df313efc0db5e63

Use the proper scaling factor to map transform block offset to the
pixel offset in the reconstruction process.

Change-Id: I9af434fbc3a0a32749dec42eebb3461567943b0f

1) Not every transform's internal signal is designed to fit in 16 bits.
2) If overflow happens in this function, it indicates that we need to
adjust the txfm's scaling. We shouldn't mute the overflow signal.
3) Saturation might be handy when all of our transform design are stable,
but I don't think we are at the stable point yet.
4) This will fix C/Trans16x16DCT.AccuracyCheck/1 failure in highbd mode.

Change-Id: I5ef5d130c22adb4b8c3b608ffcb0f2c99dc7523f

The previous refactoring CLs resolve all the potential issue
related to these two conditions. It is safe to remove them now
from the cb4x4 mode.

Change-Id: Ieeb1ca2804318c1f46b60c041538f8ca5f6ba0e7

Change the table content to support 4x4 mode_info unit. Refactor
partition_plane_context() to reflect the fact that the minimum
partition block size is 8x8.

Change-Id: Id26fcfba8d74efdb9ae316c9d066cd65358dd7e6

Replace the 4x4 block counting with minimum transform block size
counting. This retains identical coding statistics.

Change-Id: I13ce6b5b62641304b0ff1c6efd34c6a590195792

With this patch affine or homography models can be enabled
by simply changing the value of the GLOBAL_TRANS_TYPES
macro in common/mv.h to 4 and 5 respectively. Currently
it is left at supporting only rotzoom. There is a small
gain with enabling affine.

Also refactors costing to change based on the model type.

Change-Id: I46c1759de06c42c176c64ec21307ff347ddcc259

Change-Id: I3d0605a314fbd469133c96c76340396810242ada

Change-Id: I5935448b15e5a6f14932ed84451874f4acbaf95a

- For resolution > 1080p, this CL also disables upsampled
  references for best and realtime quality deadline besides
  good quality deadline.
  The following CL only disabled the good quality deadline:
  https://chromium-review.googlesource.com/#/c/405761/

- The issue was found by a segmentation fault happened in 4K
  --best mode encoding test. Current CL fixes the encoder
  segmentation fault temporarily.

- Another bug is filed:
  https://bugs.chromium.org/p/aomedia/issues/detail?id=108
  to track down the segmentation fault.

BUG=aomedia:107

Change-Id: I6f4bf14762b8e92d7b501d752167414cf37a6b26

Change-Id: Ib44c713ebcccc621d4f3b9f22e8dbb638c50ff52

The commit fixes the segfault shown in  test AV1/ActiveMapTestLarge
when build with the following config:
../../aom/configure --disable-install-docs --enable-unit-tests
--enable-debug --disable-optimizations --enable-aom-highbitdepth
--enable-experimental --enable-ans --enable-bidir-pred
--enable-dual-filter --enable-entropy --enable-ext-inter
--enable-ext-interp --enable-ext-intra --enable-ext-partition
--enable-ext-partition-types --enable-ext-refs --enable-ext-tx
--enable-loop-restoration --enable-filter-intra --enable-motion-var
--enable-new-quant --enable-ref-mv --enable-supertx --enable-var-tx

BUG=aomedia:99

Change-Id: I900d5e1d5458c482a1f0a3dde769c1ad93acdca7

Fixes a memory issue and computes values correctly for 1x1 and
1xn or nx1 videos.
This also fixes a test failure in AV1FrameSizeTestsLarge.OneByOneVideo.

Change-Id: Ib5a4efd094763fb6ceb119383c11f776784f58a4

Performance gain using baseline with ref_mv on is improved from
1.781% to 1.846%

Change-Id: I5a4bfdfcde25fa5c81e6999e20033ee4af993185

This accounts for scalable mode_info size and maximum coding block
size.

Change-Id: I82277d09100532d930e13a0a8cbfee33c99e3d64

Remove the decoder side implementation. Use the common functions
for both encoder and decoder.

Change-Id: Idcb4deedd6d2de5476b7f2e7e907fd4a7761fafa

Make the block size and mode_info size scalable to mode_info unit
size.

Change-Id: I38c49bad118e0565a104cb6edfc3e3266519f9ee

Change-Id: Ib7807125e533c32dad04b02bc54838dd1e90e8af

This is to avoid compiler warning of "mixed declarations and code".

Change-Id: I887a1cf27e777672563bda527b3d6cf70a589a8a

This commit renames a few test from SSE4_1_C_COMPARE/* to SSE4_1/* to
be consistent with the rest of the codebase and friendly to the setup
of jenkins.

Change-Id: I6b195bc0692bb6118a3dca2951d19e153792b07b

Improves results for videos that have a strong global motion.

lowres: -0.630%
(waterfall: -9.468%; tempete: -6.560%)

Change-Id: If9e7c7a5cdf12b607c5ef3392e10606789018f3b

Change-Id: Idf32db95ad25747cd6fa6fc2803b012e7994cf3c

Fixes Issue 96: https://bugs.chromium.org/p/aomedia/issues/detail?id=96&q=&desc=3

Change-Id: I47381ef3930368901c7c2ca6d7f9064216de8ad0

Replace the arithmetic calculation with direct table access.

Change-Id: I054a4c398f2321a39330aaf00607dc180a335b2d

Change-Id: I1e016bbfa3e10b78f79a4f8331de5b890fcc38d0

Separate the use cases of number of 8x8 blocks from those of
mode_info array. This allows the mode_info array processing to be
scaled to 4x4 resolution.

Change-Id: Iab78f2540355ce7658d9ea21e902a86f71148d8f

This avoids the alignment issue in the SIMD implementation.

Change-Id: I0cfe1f2ef8e1ea866e0c6f32e34c6b22ac5febd2

Change-Id: I73bcb8ab5727e2d07e34ca35e9e014f3c6f63d56

The module blocks are now scaled to operate in the unit of 2x2
blocks. This commit makes the sub8x8 intra mode RD search compatible
with such assumption.

Change-Id: If86c87c2ee011a6bc1d2de2cf2d41f270ed2896b

Scale up the mb_mode_info resolution from 8x8 to 4x4 unit.

Change-Id: I60ee20f5fc353a3c16ced14b55422fc5e4ea1e06

Change-Id: I0e1954ca281df60e56c3b1a5dcd54c47c4256649

The pixel number remains the same.

Change-Id: I40e54f4c0893db8f9452624f8167ead994b71f1d

Change-Id: I99d3abd47480ebd73fc0034fac82663c003f7bb8

Relative to previous ext-inter:
lowres: -0.177%
     or -0.029% (with USE_RECT_INTERINTRA = 0)

* When predicting interintra modes, the previous code did not provide
  the intra predictor with the correct context during rdopt. Add an
  explicit 'ctx' parameter to the relevant functions, to provide this
  context.
  This fixes a nondeterminism bug, which was causing test failures in
  *EncoderThreadTest*

* For rectangular blocks, build_intra_predictors_for_interintra needs
  to overwrite part of the context buffer in order to set up the
  correct context for intra prediction. We now restore the original
  contents afterwards.

* Add a flag to enable/disable rectangular interintra prediction;
  disabling improves encoding speed but reduces BDRATE improvement.

Change-Id: I7458c036c7f94df9ab1ba0c7efa79aeaa7e17118

Map the block offset to pixel positions in the unit of minimum
transform block size. This change retains the same coding
statistics. The encoding speed is slightly faster (<1%).

Change-Id: I2641d2b1e317ef4cdf477718c446ef025b8eef41