GitLab

Change-Id: I36f9033492f12fda7f9ce0e7a97a009aab9aab57

3840x2160

vp90-2-08-tile_1x8_frame_parallel.webm
vp90-2-08-tile_1x8.webm

vpxenc crowd_run_2160p50.y4m \
  --codec=vp9 -p 2 \
  --frame-parallel=${fpm} \
  --tile-columns=${tc} \
  --limit=10 \
  --auto-alt-ref=1 \
  --lag-in-frames=5 \
  --target-bitrate=2500

Change-Id: I6dc19b4fca483d03ef9a897a843bf9037d095c8e

The test vector exercises the loopfilter behavior at frame boundary.

blue_sky_1080p25.y4m --good --cpu-used=2 --threads=0 --profile=0
--lag-in-frames=25 --limit=300 --min-q=0 --max-q=63 --cq-level=20
--end-usage=0 --auto-alt-ref=1 -p 2 --kf-max-dist=9999 --kf-min-dist=0
--drop-frame=0 --static-thresh=0 --bias-pct=50 --minsection-pct=0
--maxsection-pct=2000 --arnr-maxframes=7 --arnr-strength=5
--arnr-type=3 --sharpness=0 --undershoot-pct=100 --target-bitrate=6000

Change-Id: Ibd0807395d2fe87f24f81f990369678df3de7c23

Also added a test vector for this issue.

Change-Id: I8a8654f9d33b27afd53c295f1ea44d198550d067

Added the test vector provided by Attila, which caught the bug in
Issue 661 "Decoder produces mismatched outputs with ssse3 enabled
and disabled"

vp90-hantro-stream-001.ivf
size: 320x180; 20 frames

Change-Id: Ic0d2b57ac7596ecb938dd55abc8c706fc2dd6d8f

Change-Id: Id578a5fe2039631cefd82dc2ef98cc62683194c3

1920x1080

vp90-2-08-tile-4x4.webm
vp90-2-08-tile-4x4.webm.md5

vpxenc blue_sky_1080p25 --frame-parallel=0 \
--tile-columns=2 --tile-rows=2 \
--lossless=0 --error-resilient=0 \
--min-q=0 --max-q=30 --kf-min-dist=0 --kf-max-dist=9999 \
--cpu-used=0 --codec=vp9 -p 2 --good --fps=25/1 --limit=5

vp90-2-08-tile-4x1.webm
vp90-2-08-tile-4x1.webm.md5

vpxenc blue_sky_1080p25 --frame-parallel=0 \
--tile-columns=0 --tile-rows=2 \
--lossless=0 --error-resilient=0 \
--min-q=0 --max-q=30 --kf-min-dist=0 --kf-max-dist=9999 \
--cpu-used=0 --codec=vp9 -p 2 --good --fps=25/1 --limit=5

Change-Id: I77d0ccf9b721ae5cfa72781fe0d3c534ea43b756

SVC multiple layer per frame encoding is invoked with vpx_svc_init and
vpx_svc_encode. These interfaces are designed to be invoked from ffmpeg.
Additional improvements:
- make dummy frame handling a bit more explicit
- fixed bug with single layer encodes
- track individual frame sizes and psnrs instead of averages
- parameterized quantizer, 16th scalefactors, more logging,
- enabled single layer encodes to generate baseline
- include new mode for 3 layer I frame with 5 total layers

Change-Id: I46cfa600d102e208c6af8acd6132e0cc25cda8d4

1920x1080

vp90-2-08-tile_1x2_frame_parallel.webm
vp90-2-08-tile_1x2.webm
vp90-2-08-tile_1x4_frame_parallel.webm
vp90-2-08-tile_1x4.webm

vpxenc blue_sky_1080p25.y4m \
  --codec=vp9 -p 2 \
  --frame-parallel=${fpm} \
  --tile-columns=${tc} \
  --limit=10 \
  --auto-alt-ref=1 \
  --lag-in-frames=5 \
  --target-bitrate=2500

Change-Id: Id9c94b722cc553a6865d443a94e8482c78b038bb

This reverts commit a82001b1, reversing
changes made to f6d870f7.

This commit breaks windows builds and needs some work to fix those and
some additional comments.

Change-Id: Ic0b0228e36704b127e5e399ce59db26182cfffe7

These changes were originally made in the Stratacaster team-review repository

commit e114bffcd82ad74c3696ec58e13c0ac895d6c82d
Author: Charles 'Buck' Krasic <ckrasic@google.com>
Date:   Mon Oct 14 16:52:13 2013 -0700

    Make dummy frame handling a bit more explicit, fixing bug
    with single layer encodes.

Squashed commit of the following:

commit 1ebbfd976c0fadb02bf1ea562a2d0e3f0206daad
Merge: ac468dde 54e88b7
Author: Ivan Maltz <ivanmaltz@google.com>
Date:   Fri Oct 11 17:29:58 2013 -0700

    Move SVC code from vp9_spatial_scalable_encoder to libvpx module accessible from ffmpeg

commit 54e88b78b160becc9569fc3c6cb6b0a8c95dc357
Author: Ivan Maltz <ivanmaltz@google.com>
Date:   Tue Oct 8 09:08:40 2013 -0700

    common svc encoding code for sample app and ffmpeg

    added svc_encodeframe.c, svc_context.h, svc_test.cc

    vp9_spatial_scalable_encoder uses vpx_svc_encode

commit 5616ec8e2e3d3e8d277333d8a9242f6c70151162
Merge: 4528014 e29137df
Author: Ivan Maltz <ivanmaltz@google.com>
Date:   Tue Oct 8 08:47:58 2013 -0700

    Merge branch 'master' into stratacaster

commit 45280148450b1f3d61e390df8aadedf85cd5bce1
Merge: bb2b675 1ab60f7b
Author: Sujeevan Rajayogam <sujee@google.com>
Date:   Fri Oct 4 10:22:31 2013 -0700

    Merge branch 'master' into stratacaster

commit bb2b675e595dc9bfc8551e963edf56800c3aea61
Author: Sujeevan Rajayogam <sujee@google.com>
Date:   Wed Oct 2 12:37:26 2013 -0700

    Track individual frame sizes and psnrs instead of averages.

commit c6d303b714795c81e7ceb4173967115c9f8ff5b7
Merge: fa87df9 35830879
Author: Sujeevan Rajayogam <sujee@google.com>
Date:   Fri Sep 27 10:05:35 2013 -0700

    Merge branch 'master' into stratacaster

commit fa87df94fba923d9f7aeb8ae20c6e15f777e00b5
Merge: bf22d71 3c465af2
Author: Sujeevan Rajayogam <sujee@google.com>
Date:   Thu Sep 26 16:10:31 2013 -0700

    Merge branch 'master' into stratacaster

commit bf22d7144895a82e0c348ac177c8a261b9e2b88e
Author: Sujeevan Rajayogam <sujee@google.com>
Date:   Thu Sep 26 11:10:34 2013 -0700

    Parameterized quantizer, 16th scalefactors, more logging, enabled single layer encodes to generate baseline.

commit ceffd7e6025b765f9886b5ea0f324248aa37e327
Author: Sujeevan Rajayogam <sujee@google.com>
Date:   Thu Sep 19 10:04:49 2013 -0700

    - Include new mode for 3 layer I frame with 5 total layers.
    - Refactor svc api.

Change-Id: Ie4d775e21e006fa597d884c59488dc999478e9b5

vp90-2-07-frame_parallel.webm:
vpxenc stefan_sif.y4m \
  --codec=vp9 -p 2 \
  --frame-parallel=1 \
  --limit=10 \
  --auto-alt-ref=1 \
  --lag-in-frames=5

Change-Id: I7381a69aaaec238b309169a51b34cb6bf29a9c50

Adding appropriate test vector vp90-2-06-bilinear.webm.

Change-Id: Ia3bbf57318e0cc61a1b724fe751e3f9c7e11b337

Modified the resize unit test so that it optionally
writes the encoded bitstream to file. The macro
WRITE_COMPRESSED_STREAM should be set to 1 to enable
output of the test bitstream; it is set to 0 by default.

Change-Id: I7d436b1942f935da97db6d84574a98d379f57fb1

Added the resize_test unit test to the VP9 set.

Set g_in_frames = 0 to avoid a problem when the total
number of frames being encoded is smaller than
g_in_frames. In this case the test will not have
access to the encoded frames and will not be able to
compare them for testing for encoder/decoder mismatch.

Change-Id: I0d2ff8ef058de7002c5faa894ed6ea794d5f900b

This is a bitstream change but no currently produces videos should
be affected. https://code.google.com/p/webm/issues/detail?id=610

Change-Id: Ic85a6477df6c201cdf7f70f6bd84607b71f4593c

This adds a test vector for 444 color space.

Change-Id: I1e2ac3883211989a062cfafc0e58151b14d294b8

vp9/decoder/vp9_thread.[hc]
Original source:
 http://git.chromium.org/webm/libwebp.git
 100644 blob b1615d0fb8d311666b2fa4561076c62d72c2e3ff  src/utils/thread.c
 100644 blob 13a61a4c84194c3374080cbf03d881d3cd6af40d  src/utils/thread.h

Local modifications:
 - s/WebP/VP9/g
 - camelcase functions -> lower with _'s

Change-Id: Ib6932640ee34f8b4782c6fbd15864a59d5d4c5fe

Adding missed parenthesis around boolean expressions. Bitstream is changed.
Regenerating test vectors.

Change-Id: I4cc00b761e9473f92f180a9fc3a0c607f0aaae56

It does encodings with min and max q set at 0, and check to make sure
output PSNR at MAX_PSNR (100).

Change-Id: Ia2418353cccf6e487204ea4ff874a7e71e55cb3e

In the rare case were 4x4 interior filtering was called for but no
8x8 or larger filtering takes place, the previous code was skipping
the filtering. This patch fixes the issue by including the interior
mask in the overall mask for the filter application loops.

Change-Id: I4a0b65056c64f97478827c2ff41e0914fc7779d0

Tests q0 ( lossless),  very high bitrate and low bitrates at cpu speed
0, 1 and 2.

Change-Id: I0c5cdca00acd8d01e7b13f124b3b08d4b1ae9f6d

For cases where there's no transform set in bit 0 (the left edge of
the SB) but bit 0 of mask_4x4_int is set (the edge 4 pixels from the
left edge needs filtering), it was incorrectly being skipped before.
This situation only happens on the leftmost edge of the image, as
the edge at column 0 is intentionally skipped since there aren't
pixels to the left to read.

Change-Id: Ib2fbbcb40166e90af31b1a0e13b85b68c226cbd3

3% faster overall (3min35.0 to 3min28.5).

Change-Id: I5ff8a5c2c91586b6632ca5009ad1ea51ce94af5e

Tests resolutions of 8, 10, 16, 18, 32, 34, 64, 66 to exercise the
border conditions, as well as non-SB aligned sizes.

Change-Id: Ie7c2b7860ac3727e23202042f2e86792652912f8

add ClearSystemState() to reset MMX registers avoiding corrupting
subsequent tests.

Change-Id: I668deb09aa7aa467709776e5819f936910698bc0

These files can stand in until we get proper syntax vectors. They
should provide some additional assurance against inadvertant
bitstream changes.

Change-Id: I12f6c9a5f054e30df40a7ff1f33145abf7e1d59d

This avoids encoding tokens for blocks that are entirely
in the UMV border. This changes the bitstream.

Change-Id: I32b4df46ac8a990d0c37cee92fd34f8ddd4fb6c9

New decoder control paramter VP8_SET_DECRYPT_KEY to set the decryption key.

Change-Id: I6fc1f44d41f74f3b3f702778af1a6f8f5cc9439f

The commit changed the name of files and function to remove obselete
reference to LLM and x8.

Change-Id: I973b20fc1a55149ed68b5408b3874768e6f88516

A 'superframe' is a group of frames that share the same PTS, but have a
defined decoding order. This commit adds the ability to append an index
to such a group of frames, allowing for random access to the constituent
frames. This could be useful for frame-level parallelism or partial
decoding in a multilayer scenario.

Decoding the stream serially without such an index should work as a
fallback, and VP9/TestSuperframeIndexIsOptional verifies that.

Change-Id: Idff83b7560e1a7077d8fb067bfbc45b567e78b1c

Change-Id: I06b5ba5c457944cfa4cd9f53c3bd8cda132439c2

Also
1. Removed the test code for fDCT from the iDCT test.
2. changed the criteria of round trip error to be below 1/block, this
is quite strict comparing to smaller transforms when size differences
are accounted for.

Change-Id: Idb46a6380b04c93fc8e2845c75f5a850366b0090

This commit added pre/post scaling for first half of fDCT16x16 to
reduce error, by simulation of 100,000 blocks for random inputs,
the average sse reduced from 2.1/block to 0.0498/block.

also enabled tests for 16x16 fDCT and iDCT

Change-Id: Id2a95f0464c6dd4118797d456237ae90274c0f02

The commit added a final rounding choice for 8x8 forward dct to get
rid of a sign bias at DC position and improve the accuracry in term
of round trip error for 8x8 fDCT/iDCT.

This commit also enabled forward 8x8 dct test.

Change-Id: Ib67f99b0a24d513e230c7812bc04569d472fdc50

Change-Id: I4e94ee2c4e2360d6a11a454c323f2899c1bb6f72

1. changed 4x4 test name to Vp9Fdct4x4Test to be consistent
2. remove forward 8x8 dct test code from idct8x8_test.cc
3. temporarily disable other forward dct tests to allow fdct work in
progress

Change-Id: I566aeed9c7c34da5a206190aa7d0e847a4008b36

This patch adds column-based tiling. The idea is to make each tile
independently decodable (after reading the common frame header) and
also independendly encodable (minus within-frame cost adjustments in
the RD loop) to speed-up hardware & software en/decoders if they used
multi-threading. Column-based tiling has the added advantage (over
other tiling methods) that it minimizes realtime use-case latency,
since all threads can start encoding data as soon as the first SB-row
worth of data is available to the encoder.

There is some test code that does random tile ordering in the decoder,
to confirm that each tile is indeed independently decodable from other
tiles in the same frame. At tile edges, all contexts assume default
values (i.e. 0, 0 motion vector, no coefficients, DC intra4x4 mode),
and motion vector search and ordering do not cross tiles in the same
frame.
t log

Tile independence is not maintained between frames ATM, i.e. tile 0 of
frame 1 is free to use motion vectors that point into any tile of frame
0. We support 1 (i.e. no tiling), 2 or 4 column-tiles.

The loopfilter crosses tile boundaries. I discussed this briefly with Aki
and he says that's OK. An in-loop loopfilter would need to do some sync
between tile threads, but that shouldn't be a big issue.

Resuls: with tiling disabled, we go up slightly because of improved edge
use in the intra4x4 prediction. With 2 tiles, we lose about ~1% on derf,
~0.35% on HD and ~0.55% on STD/HD. With 4 tiles, we lose another ~1.5%
on derf ~0.77% on HD and ~0.85% on STD/HD. Most of this loss is
concentrated in the low-bitrate end of clips, and most of it is because
of the loss of edges at tile boundaries and the resulting loss of intra
predictors.

TODO:
- more tiles (perhaps allow row-based tiling also, and max. 8 tiles)?
- maybe optionally (for EC purposes), motion vectors themselves
  should not cross tile edges, or we should emulate such borders as
  if they were off-frame, to limit error propagation to within one
  tile only. This doesn't have to be the default behaviour but could
  be an optional bitstream flag.

Change-Id: I5951c3a0742a767b20bc9fb5af685d9892c2c96f

This commit introduces a new convolution function which will be used to
replace the existing subpixel interpolation functions. It is much the
same as the existing functions, but allows for changing the filter
kernel on a per-pixel basis, and doesn't bake in knowledge of the
filter to be applied or the size of the resulting block into the
function name.

Replacing the existing subpel filters will come in a later commit.

Change-Id: Ic9a5615f2f456cb77f96741856fc650d6d78bb91

test for forward transform will be enabled later after re-do forward
transform

Change-Id: Ie7c7cf88baf7ecbebbe52fe027e1c3b33d3b9d49