GitLab

Change-Id: Ied1a493723cac2ec4674bac642525a937826733f

Change-Id: I802c9e41ebfed090b5ad8300917aad5e16ad026a

Added: CONFIG_DAALA_DCT, CONFIG_GF_GROUPS, CONFIG_MRC_TX

All default to 0, off.

Change-Id: Ic7f8a6f68038b6138b64a611e91315212605f0c3

And correct SO_VERSION_MAJOR in libs.mk.

BUG=aomedia:627

Change-Id: Ia0a3bb5774fa44c182e4f8072ffe90970ee2aeb5

The post relocation destination of some non-simd sources
that require specific simd flavors was incorrect. The
relocated sources touch internal APIs, and the incorrect
destination was an unguarded area where sources are added
unconditionally to targets with the assumption that all
APIs are accessible.

Sources moved to section guarded by BUILD_SHARED_LIBS, which
acts as the border between public APIs always available for
testing and those that require a static lib that exposes
private symbols.

Change-Id: I5d9ad513a0fe1b9dc57cb11cca53e8d00eb9b1a8

A failing 'assert' aborts the execution of the test suite,
instead of simply failing the erroneous test, and continuing with other tests.

Change-Id: I3fdaf20a22ea343ac75be7883addbe6ba1e55da1

Change-Id: I5d53689083472e1ff312f60f056b93a233e8770e

Currently, lgt cannot work with pvq, so those lines are not
necessary.

Change-Id: Ida7b2de599b55d94cac57365c09ac2330abe3c66

An ncobmc mode is selected for each squared block
(if motion_mode == ncobmc_adapt_weight) as opposed
to four in training

Change-Id: I81e5dfca3a6fec5a95c96b17f294530880cf8f88

Change-Id: I4b4272c99ba6ff4d983dcb694b3ea4d5de1b4e14

Added a sample selection process in warped motion.
1. Gather more samples including multiple rows on the top, multiple
columns on the left, and the upper-right block.
2. Sort samples by the MV difference between the neighbour's MV and
the current block's MV. Trim the samples with considerably large MV
difference.

Borg test result:
             avg_psnr ovr_psnr ssim
cam_lowres:  -0.241   -0.243  -0.376
lowres:      -0.104   -0.110  -0.179

The changes are wrapped in WARPED_MOTION_SORT_SAMPLES macro.

Change-Id: I2730bb31a0a3ad28215ccd16fd6da0ea8b2ed404

The memset to 0 wasn't required because the temporary variable is only
written to, before being read in the next function call.

Tested:
./test_libaom --gtest_filter=*SelfguidedFilterTest*

Change-Id: Ie1628d43b050744ae97a8be55f551edb602b018b

Change-Id: I2888bd8905253e02e3ac74597275cf56e5142d29

Always let the encoder decide if "allow_screen_content_tools" should be set.

-The screen content detector(is_screen_content()) works fairly well.

-Keeping the config control may lead to mismatches. For example, user
may encode the 1st frame with the flag on, then turn the flag off for
the 2nd frame via config control. On the decoder side, the flag is on
for both frames, because the flag is only transmitted in the header of
all-intra frame, i.e. the 1st frame in this example.

BUT=aomedia:656

Change-Id: I0652fe2b9f8df19bfbf7ed7d3513205c80ea9c61

Change-Id: Ib245b059f1c1aa450b0755b437eab15c362d3aee

Files that contain a simd flavor in their name are only instrinsics
sources when the simd flavor is the suffix of the basename.

BUG=aomedia:657

Change-Id: I87d47c77750730171817f38fe3877226e7c09c10

Avoid consuming error messages like "Illegal instruction".

BUG=aomedia:657

Change-Id: Ia7983389ad3519f795f85d0c94a2a3787614a28d

Change-Id: Ie52ce24ca9eca272dfacfb17483ec7b7f2c126d4

This is a partial fix of the HBD upsampled pred quality loss
introduced in commit 5d24b6f0.

BUG=aomedia:649

Change-Id: I114dab1c8aac285071c5910dff40d48beea8c2ed

This simplifies the path from rd_pick_intra_sbuv_mode()

Results on Subset1 (compared to  dff41923

 with CfL enabled)

  PSNR | PSNR Cb | PSNR Cr | PSNR HVS |   SSIM | MS SSIM | CIEDE 2000
0.0000 |  0.0000 |  0.0000 |   0.0000 | 0.0000 |  0.0000 |     0.0000

Change-Id: I0bade9d347c626a78ba7077b960afdb318ecca69
Signed-off-by: David Michael Barr <b@rr-dav.id.au>

Added a 1-bit flag 'large_scale_tile'. If it is 0 that is the default value,
use normal tile coding in TILE_GROUPS. If it is 1, use large-scale tile
coding in EXT_TILE.

At large_scale_tile=1 case, if single-tile-decoding is required, then the
loopfilter is disabled.

Related API and unit tests were modified.

Change-Id: I3ba12dc3d80ccf1ab21543ab3b16c02282c34e3b

Change-Id: If878c33bb46c13216bd958c906f69d977066a144

Change-Id: I2d688d21de4a1d180f8ac36bb0363b57be55c0af

Transmit ncobmc modes only when the motion mode,
ncobmc_adapt_weight, is selected.

Change-Id: I19d6dd83278457b1867f8e2e5cd94d25f535fcaf

Change-Id: Iac17a7069749f10842dec75d5791aeadfa18f0f5

Before this, this function was copied-and-edited from
av1_build_inter_predictors_sby() and av1_build_inter_predictors_sbuv().
Also, some places used sb() one directly (like rdopt.c), while others used
sby() and sbuv() separately (like encodeframe.c --> encode_superblock()).

This was prone to errors, if someone mistakenly modified only one/two of
the 3 functions.

Now, we just reuse sby() and sbuv() functions in sb().

Verified that there's no change in output.

Change-Id: If81d78e767cc3fa09f06094d4db36ff1eaea3e4f

Change-Id: Iffc8e04bb5fa5c63542d5f85e838e6bc60405099

BUG=aomedia:654

Change-Id: Ic77a456246f87acb59ae95ddd4c421e741a99789

With aq-mode=1 (which forces segmentation), AWCY
gives -0.12% (objective-1-fast, cpu-used=4,
NEW_MULTISYMBOL on for reference and tested).

No change to baseline metrics, with or without
NEW_MULTISYMBOL.

Change-Id: I841a9e8978598402dacfbca9265269163f7440c0

Use row and column indexes to fetch the txk_type value.

Change-Id: I881d500c030e322d8aca9dccb6ff2870c9e1e392

Change-Id: Id89b613a227867782f9feba16d87eca4b5a86369

The goal is:
1) to provide a tool allowing to diagnose
and fix mismatches between LBD/HBD coding paths.

2) once all mismatches are fixed, this test
should be enabled so new ones can't be introduced.

Change-Id: I71c1db4c2cb142e9e3a7a4b02dd0d61f35b77178

- Decoder convolve rounding cycle percentage drops from
  2.75% to 0.91% by using avx2 function on i7-6700.

Change-Id: I34ae48f45c0b4073f8962647d2181365ffe3325b

Due to rounding the variance can become negative.
Similar behaviour has been fixed in commit b7e7c044.
I have made the corresponding changes to the masked_variance code.

BUG=aomedia:646

Change-Id: I21ed5a3f58080def08ac7ab9a77cbcad90318c99

Change-Id: Ia75a5160fc11df5d5dcbcb1bcf767de9925f811e

This commit resolves the compression performance regression when
both compound rounding and compound segment are enabled.

Change-Id: I831b4e44d2b054aa74922eb0c541eaa2ac355e0e

Use multi-symbol CDF coding instead.

Change-Id: I8a69364c8672474021d2ea5933e344d493186967

The frame marked as last_bipred at the encoder side is the frame right
before the source frame of BWDREF_FRAME, which was not used as a
reference. This CL is to add this type of frames to the reference frame
buffer and use them as reference for the coding of future frames. This
is a non-normative change.

A tiny gain is observed (lowres BDRate -0.016% in avg psnr). The main
target of this CL is to make it possible to have more flexible gf group
structure. For instance, LAST_BIPRED_UPDATE can be replaced by
LAST_UPDATE followed by an extra alt ref frame.

Change-Id: If3dbce0959bb468e95d01a1feca1dac6bfb91d36

Add a new default motion_mode_cdf supporting this
experiment.
Also define the costs for the ncobmc mode tokens.

Change-Id: Iebab044e4a34e3d2a94c18893107d82055cc9c77

Change-Id: I89035a64414f9511789022704ad9343bc2937666