GitLab

Change-Id: I1ce575377c3d971f4488177b82c6e0b32aa5e4be

Reduce the motion interpolation filter taps for inter prediction
blocks with widths or heights smaller than or equal to 4 to alleviate the memory
bandwidth increase.

AWCY HL: 0.01% Y, -0.20% U, -0.29% V (positive number means loss)

Change-Id: Ic454340e20aea2f1aae622336990f24a9e5b54d8

This experiment uses switchable interpolation filter for intra
prediction. Current gain is too small to be useful for AV1.
May explore more for AV2 in the future.

Change-Id: I24ae8c75dc71adf6271eabdb68a81b68a0869fa2

Change-Id: I22aed63c6659a7301bccaa994b7374af548db5d1

Change-Id: Ia9a5d818e8c2ff9b4cc41c6d7950cfe005c20bfc

Change-Id: If8f8e1a0032e914beb3ec3bcde221fe4a5605139

Temporarily disable the use of 10/12-tap filter.

The filter type coding dependency on motion vector coding has been
removed previously. Every inter block now has to send filter type
regardless of the motion vector position.

Enable the dual filter scheme.

Change-Id: I81086f5f9de6375ad73bb4120be0c87b341120b1

SIMD convovle functions, such as filter_horiz_v4p_ssse3(), assume that
10-tap filters are defined using 12 taps with both end taps being 0.

BUG=aomedia:380

Change-Id: Id8a87ae8a1330bed0452441ab8345276857220af

Change-Id: I10cb9af13d50f2e5ac471af86a4393739c28b509

Performance drop
lowres 0.056%
midres 0.024%
hdres 0.02%

Change-Id: I52d067eefbfb87198319f9d50e3b4060f80a6abb

So far the overall performance drop by
lowres: 0.027%
midres: 0.049%
hdres: roughly 0.07%

Change-Id: If3dc7619d91cb97b6d494b33f36112af45efcd8f

This allows Y and UV planes use different filter kernel

Change-Id: I524aa3f6ffbc439acee6660bdbaf5cf2059379b5

Change-Id: I1bf9c2fcd664fb4012d67e420cac613eaca735a8

Change-Id: I0ebd6951d2b42869ae872b33f63a07db03e99c62

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

Performance drop
BDRate
lowres -0.116%
midres -0.073%
hdres  -0.056%

Change-Id: Ic90caf9b8f6fb9d9fd6f9c0e80436a7c468a3c97

Compression will drop slightly
       BDRate
lowres -0.096%
midres -0.069%
hdres  -0.084%

However, this will reduce the combinations of filter types needed to search in
dual_filter experiment

Change-Id: I75d4ddc73b91099e17fd52a1b1174bd3d2ab3e18

Purpose:
-Reduce dynamic range of interpolation filter coefficents from 8
bits to 7 bits.
-Inner product for 8-bit input data can be stored in a 16-bit signed
integer.

Impact on compression efficiency:
-Marginal improvement, typically less than 0.5% BDR.

Change-Id: I58d1408307ae7d2a6f9de8965c5877b258703199

This is a similar change to following aom CL
https://aomedia-review.googlesource.com/#/c/1961/

Move SIMD related functions from filter.c/h to following files
av1_convolve_ssse3.c
av1_highbd_convolve_filters_sse4.c

Change following c files to header files.
av1_highbd_convolve_filters_sse4.c
av1_convolve_filters_ssse3.c

Change-Id: I41a3cc6b0789e632451aeda82f5eb97a4d78e370

- Move filter look-up functions to corresponding optimization modules.

BUG=webm:1296

Change-Id: I87f399609052db2dbc7e5a590afb08b82e3fa89f

Purpose:
-Reduce dynamic range of interpolation filter coefficents from 8
bits to 7 bits.
-Inner product for 8-bit input data can be stored in a 16-bit signed
integer.

Impact on compression efficiency:
-Marginal improvement, typically less than 0.5% BDR.

Change-Id: I58d1408307ae7d2a6f9de8965c5877b258703199

this matches style guidelines and stabilizes successive runs of
clang-format across the tree. remaining types should be address in
successive commits.

Change-Id: I6ad3f69cf0a22cb9a9b895b272195f891f71170f

Change-Id: I2b2b70e756b7eb9611b7b33b7d5f19b3b30e0a50

Cherry-Picked the following commits:
0defd8f2 Changed "WebM" to "AOMedia" & "webm" to "aomedia"
54e66767 Replace "VPx" by "AVx"
5082a369 Change "Vpx" to "Avx"
7df44f17 Replace "Vp9" w/ "Av1"
967f722f Remove kVp9CodecId
828f30ce Change "Vp8" to "AOM"
030b5ffc AUTHORS regenerated
2524caee Add ref-mv experimental flag
016762be Change copyright notice to AOMedia form
81e55269 Replace vp9 w/ av1
9b94565b Add missing files
fa8ca9f2 Change "vp9" to "av1"
ec838b76  Convert "vp8" to "aom"
80edfa01 Change "VP9" to "AV1"
d1a11fb9 Change "vp8" to "aom"
7b582513 Point to WebM test data
dd1a5c8d Replace "VP8" with "AOM"
ff00fc0f Change "VPX" to "AOM"
01dee0bb Change "vp10" to "av1" in source code
cebe6f0c Convert "vpx" to "aom"
17b05679 rename vp10*.mk to av1_*.mk
fe5f8a8a rename files vp10_* to av1_*

Change-Id: I6fc3d18eb11fc171e46140c836ad5339cf6c9419

Manually cherry-picked commits:
ceef0583 libvpx->libaom part2
3d26d916 libvpx -> libaom
cfea7dd7 vp10/ -> av1/
3a8eff73 Fix a build issue for a test
bf4202ed Rename vpx to aom

Change-Id: I1b0eb5a40796e3aaf41c58984b4229a439a597dc

clang-format authored Aug 11, 2016 and James Zern committed Aug 12, 2016

Change-Id: I01d8241eba3ccaf4d06c00a51df2d17c126f6f9d

- For experiment EXT_INTERP under high bit depth.
- Add unit test to verify bit-exact.
- Speed performance improvement:
  On Xeon E5-2680, park_joy_1080p_12.y4m, 50 frames, encoding time
  drops from 6682503 ms to 5390270 ms.

Change-Id: Iea4debf5414f3accf1eb5672abeab56a0539ac77

- Apply signal direction/4-pixel vertical/8-pixel vertical
  parallelism.
- Add unit test to verify the bit exact result.
- Overall encoding time improves ~24% on Xeon E5-2680 CPU.

Change-Id: I104dcbfd43451476fee1f94cd16ca5f965878e59

Remove the unused sub-experiments within the ext-interp experiment.

Change-Id: I716e3392412d02c56f9395a86c9cab02f580fa59

Change-Id: Ia055d03656ad1580447eced8687949583fdf4089

There is a ~0.1% gain.

Various experiments with different kinds of windowing functions to
follow.

Change-Id: I0787fddca53607ab39e53f919066839301938e68

Fixes some issues introduced by a merge of two patches.
Also decouples the temporal interpolation filter from the switchable
filters for now for ease of experimentation with both separately.

Change-Id: If1c7c08adf00e0cf818fe8d0d3656c26ea65eb32

Includes various cosmetic changes and refactoring including
naming the sharp filters differently (since they are no longer
8-tap).

Change-Id: Ida5a19ca0daa9f6a64a6734394c685b2a4a2564a

This commit uses 12-tap sharp filter to generate alter reference
frame. It improves the compression performance by
derf    0.45%
hevcmr  0.35%
stdhd   0.79%

No encoding time change is observed.

Change-Id: Ia5dc26d5aae6b9b0cb782e5a28dc5066eeeb2ec8

Change-Id: I7df48e7f3b57f212798ef4be86f28aed928fc3e0

-Avoid unnecessary calculations
-Use SIMD when possible

Encoder is about 5% faster with the extra intra prediction angles
enabled.

Change-Id: I131056befe327cedab217ad4a40d5f2a11318acc

Set USE_12_SHARP_FILTER to 1 to turn on the experiment
The psnr percentages increase
       derf   stdhd
lowbd  +0.332 +0.318
highbd +0.476 +0.507

Change-Id: I783c0fc764ee8541645e100453c9b2073924e209

Change-Id: I1a5b1e15a188ef11594d0c6ac0dbd42aac59cfca

BD-rate performance improvement (on top of ext-intra):
derflr  0.22%
hevclr  0.36%
hevcmr  0.48%
hevchr  0.37%
stdhd   0.19%

Average speed impact on some derf clips is about 40% slower (on
top of ext-intra). Speed improvment is a to-do.

Change-Id: I8fe3fe8c5e4f60d0462778adbcc15c84dfbe7a25

clang-format authored Jan 27, 2016 and James Zern committed Jan 28, 2016

Change-Id: Ib8c9eb6263d6eba6b9d7b2e402b7e83a78c86be9