GitLab

Even though number of unique colors in a block may be at least
PALETTE_MIN_SIZE, sometimes the K-means can generate fewer than
PALETTE_MIN_SIZE unique centroids due to integer rounding.

This was resulting in "k - PALETTE_MIN_SIZE" being negative and using
that as an index into palette_y_size_cost array was causing a UBSan
error.

BUG=aomedia:515

Change-Id: Ia00b4bb2efebbe43d1fec7b619ebb837b1ae961c

Change-Id: I64c523a34b4d69450759433e125a79e63d928037

math.h provides these with _USE_MATH_DEFINES

BUG=aomedia:500

Change-Id: I6e05b09149a6a6ca152e67194b629e0728bc3c35

add correct type for 'c' parameter

BUG=aomedia:500

Change-Id: I4ea2334405d8d7ea436bc373dd1c812666f98067

Measure SSE for all possible alphas.
Estimate rates for alpha signalling.

Change-Id: Idf1e3c632925cd306090fc38cf5b95eff7ee5c1c
Signed-off-by: David Michael Barr <b@rr-dav.id.au>

Change-Id: I7b26893e9358378f585a892854812ed0a1888385

The small multiplier intentionally introduces a larger error, so
relax the accuracy threshold in the aom_reader_tell_frac() test.

Change-Id: Ib046a95dd3a8498a1a2ea58680796e75fd195868

Change-Id: Iad3c54cc905a7d337766c3c8ae59109327d5c5ae

Change-Id: Ib6f8b94c82fee8546d741f4472f42c342cbdd55c

BUG=aomedia:514

Change-Id: I97e20524b091169635395098efe234a7fd846bd9

'const_above_row' was same as 'above_row' and 'above_row' pointer was
never modified.

Change-Id: Ib0c2524b1afc75967cfa7d3da8e59cfc66299ff4

Change-Id: I6f0c93fd38abc85553b81a3f2d99e97ed9c803f7

Change-Id: Id490c57df5dca13c59a62b0a542779784842944c

The CDF initialization needs the iCDF macro wrapper.

BUG=aomedia:517

Change-Id: I2dd7c43ee2765f97c2fb247c31440e79587723fe

BUG=aomedia:516

Change-Id: I0af6e5e20f785bd4862f499b8d53002f58677c40

Adds a speed feature to control which references to use
to compute global motion.
Also adds logic to not compute duplicate sets of
parameters when reference frames point to the same
buffers.
Includes some renaming of functions to set good speed
features to make things clearer.

Change-Id: I641d33441fde98af18cad8d4db49cf7d5d153ead

Intra weights start from a higher value, and decay faster
Improve interintra by 0.1%/0.17% on lowres/midres

Change-Id: Ib20f47d1e89f729731e3bcfceaa21a47c8b45c56

fixes a couple of type conversion warnings for msvc2015

Change-Id: I004f072661c981e204bd62b2740201658ad67294

The conversion of return value to int64_t appears to be unnecessary.

fixes a number of type conversion warnings for MSVC2015

Change-Id: Ibca24468e73bf22ad07c469ce5d1223635390ecb

Removed "const" from parameters that are passed by value, also fixes
msvc2015 compiler warnings.

Change-Id: I0c7555b4245fa55fabda1c27762fb9f0860bfa18

Move the definition of several ssse3 only functions into a separate
header file.

BUG=aomedia:500

Change-Id: Idbaedfe23d439b6092d46f0a2aa127efebe4d59b

Change-Id: Ib92f54480035a85ab7279ffa9f8b9237d0a3f334

Change-Id: Idc37f9babf5fb061e0a36bf1ef8649af4520f307

Use a simpler least-squares function for affine and rotzoom
model estimation, instead of computing the pseudo inverse.
Also refactors the code into a separate mathutils.h file.

The SVD code is currently used only for estimation of the
homography models which can be removed when we remove the
homography models.

Coding efficiency change is in noise range, with the small
difference coming from numerical precision issues.

Change-Id: I0a9eb79495911cea21a7945b397d596e22a2a186

The angle estimation in ext-intra didn't account for
alt-intra.

Also use c99 variable declaration style.

BUG=aomedia:444

Change-Id: I001b0278eb8cf560975c873a423de024c10d08b4

This fixes a mismatch which occurs when global/warped motion and
a masked compound type are used together.

Change-Id: I08b2702cdb3b85f8d8817b9286a73951c97cf379

Writes and reads alpha to and from the bitstream.

A special case is needed on the encoder side to handle prediction block
skips. Since whether or not a prediction block is skipped during CfL, a
rollback is required if the block was skipped and the alpha index was
not zero. The advantage of this is that no signaling is required when
the prediction block is skipped as it is assumed tha the alpha index is
zero.

A encode facade is added to the intra prediction facade as CfL requires
special encoder side operations.

Change-Id: Ic3b11d0fdbd51389d862112eb09d8785127a6b06

Change-Id: I705bbb79104eda4a7db8146e49795dbe80a31a3f

Change-Id: Icde963dfa58ebea490dd1e4d125cd9ca5d27e7db

fixes MSVC build errors.
BUG=aomedia:500

Change-Id: I677b60e3b2b9b655193773f8f90a04e360e35990

this change enables loopfiltering_across_tiles by defualt after
it has been adopted. support in the deblocking filter and CDEF
filter has been implemented. a command line argument --tile-loopfilter
has beed added to enable (1) or disable (0) loop filter across
tile boundarys.

Change-Id: I6e55c32c35f23c2aedf6b6dfdab9238750b0cf34

Used CHECK_MEM_ERROR in cpi->mbmi_ext_base memory allocation.

Change-Id: Ief8629d2b420b073f233c7c8e5e0b6d4269aea5c

Reduced the test clip's resolution while testing on 32-bit system.

BUG=aomedia:510

Change-Id: Icea7a9d3affe1ea9e0842424db72bdc5bdf91390

Change-Id: I96e386678f6227f3ca735c9be2725cb9795b3296

Remove the unused codes under the experimental flag.

Change-Id: I4e1c176d13a58f6677f32d52bbfaa7f0aad4249d

The aom_codec_stream_info struct had an `sz` field. It looks like
this was intended to implement polymorphism, where individual
codecs under the aom API could extend the struct with additional
members while maintaining common fields like frame size.

However, no codec-specific initializer was ever provided,
requiring callers to manually set this field to the expected
value. The extension mechanism isn't used by the av1 codec,
and also wasn't used by the vp8 or vp9 codecs in the libvpx
implementation.

Since it's not used, this commit removes the layer of indirection,
removing the confusing initialization burden and using
aom_codec_stream_info_t directly within the av1 decoder.

Change-Id: I4372a286b1d03518345c83b01c2f51db6cefe211

Change-Id: Ibf57c631b469b1dddafb60be2cbb02d221f378d0

Change-Id: I8d104af32f2400f3133b0dd1e7cabd0b7e3e31aa

- This change comes from VP9 code review.

Change-Id: I30449e641c24fc4b1ad01595cdc4a605bc797e83

This will guarantee that av1_optimize_b will be turned off when
lossless mode is on

Remove heuristic lossless check in optimize_b_greedy

Change-Id: I636c776f3f6b632eb03bc57a470ea43aae4fe0f6