GitLab

Improves keyframe coding by 0.1% on the screen_content testset.

Change-Id: I5793a67eaae21010ef200038af99ebb9029fc770

Properly count the base range coefficient coding in the rate
distortion optimization and soft quantization process.

Change-Id: I860001f51c4a9d0021d08b85b8ccdb097121b287

Add sse4_1 functions for Intra-edge experiment:
  av1_filter_intra_edge_sse4_1()
  av1_filter_intra_edge_high_sse4_1()

Approx cycle reduction at qp 20, 1 kf:
  Enc (lbd) 1.4% to 0.3%
  Dec (lbd) 0.4% to 0.1%
  Enc (hbd) 1.1% to 0.2%
  Dec (hbd) 0.6% to 0.1%

No change to bitstream

Change-Id: I176b2d125424d7d226114c807915c33dde5c3720

This will let the coding performance drop by 0.1%,
but it will speed up encoder by 40%

The change is under the flag FAST_OPTIMIZE_TXB

Change-Id: I07dea24e818123e27c4347189aaec59deb42903c

Change-Id: Ia4ec99ce8f3dd0098764571777e78adac21b6e67

Change-Id: Iaf93f37d8b6e2e8d256f8ff5a408d45802a2ae57

- Add aom_scale dspr2 sources to the correct target (aom).
- Fix an inverted high bit depth condition.
- Remove claims that dspr2 variants of av1_iht16x16_256_add_dspr2,
  av1_iht8x8_64_add_dspr2, av1_iht4x4_16_add_dspr2 from
  av1_rtcd_defs.pl exist in low bit depth configs.

Change-Id: Ibdd42e475b81c2491f02ba10ca0d461f7ff15bc5

Change-Id: Iadbb91ca1905318f094fdfae95f2c4db87943842

Refactors and adds superres-mode 3 and associated
paramters --superres-qthresh and --superres-kf-qthresh
that are used to trigger superres mode when the qindex
for any frame exceeds the thresholds provided for non-key
and key-frames respenctively. The superres scale factor
numerator is progressively reduced from 16 starting from
that q threshold following a fixed slope.

Change-Id: If1c782993667a6fbaaa01bbde77c4924008c0d28

This patch fixes a regression introduced in 1d190950 where the encoder
 was using the 16x16 VP9/AV1 transforms for RDO, but then used the Daala
 transforms for encoding.

subset1:

master-daala_dct16@2017-09-13T12:05:18.013Z ->
  master_daala_dct16_use_c@2017-09-13T13:05:02.252Z

   PSNR | PSNR Cb | PSNR Cr | PSNR HVS |    SSIM | MS SSIM | CIEDE 2000
-0.3654 | -0.7634 | -0.7407 |  -0.4884 | -0.4699 | -0.4945 |    -0.5104

master-no_rect_tx-no_var_tx@2017-09-12T00:23:18.153Z ->
  master_daala_dct16_use_c@2017-09-13T13:05:02.252Z

   PSNR | PSNR Cb | PSNR Cr | PSNR HVS |    SSIM | MS SSIM | CIEDE 2000
-0.0133 |  0.1040 | -0.0440 |  -0.0492 | -0.0151 | -0.0120 |     0.0699

Change-Id: Id1830d0975db4bd0320a47fdf45b4bca20881cfb

Change-Id: I0b6a1781ab3bf8e32874fe71e24a4cb8c9943357

Change-Id: I408dbcf69b3f680e5de570e080ce39aeacbf618d

This is a manual merge from libvpx's commit:
"05ee24149 Add high bitdepth intra prediction
optimization speed test". Credit goes to Linfeng Zhang.

Change-Id: Ie254593aa9b601889ecb95eca900365055d46a03

Change-Id: I9b92caa9d65a3b062762c270c89018c8e08eba0f

DISABLE_TRELLISQ_SEARCH will incur failure of TestResizeCspWorks
BUG=aomedia:734

Change-Id: I70d51b9a490b251ebd7743faf831da54b94e48c7

Change-Id: Iad6a16d3ab819405de037d5100060710e0636176

Change-Id: Ib19d9431124b2416abe4a217138ec8bdb029f73c

This change fixes a compile error when all three of --enable-daala_dct4,
 --enable-daala_dct8 and --enable-daala_dct16 are enabled at once.

Change-Id: I4942e09fb887afbda2eda6aaacec727b5cbf6f50

This experiment has been agreed to remove its flag in
Sep 12 AOMedia Codec WG meeting.
We first soft-enable it until it gets stable.

The provided encoder option "--tune=[cdef-dist | daala-dist]" is turned
off by default, hence there is no change in bd-rate.

Change-Id: Ib4553125bbaeff5a98542b39cba942496337f901

Also fix asan warnings.

Change-Id: I21fe3150635540abacb5b1b8928c6470db605035

Angie Chiang authored Sep 13, 2017 and James Zern committed Sep 15, 2017

- make Check() void as the EXPECT's are sufficient to document failure

cumulatively this has the effect of avoiding reporting incorrect Check()
failures due to earlier test failures.

This CL is ported over from
f8c27d164 register_state_check: simplify Check() methods

Change-Id: I1b65aa769c69c2a52b2e0b363f1c4432965ee89f

Add a macro to improve the readibility of test case. The
coming test cases varying on mode/size would expand the
list too large.

Change-Id: I74171344098820b21090dd9b857229bdf2e77248

Currently the use of the new multi-layer GF group of 16 is guarded by
the macro USE_GF16_MULTI_LAYER, which is turned off. Hence this CL
does not change any of the encoder behavior.

Change-Id: I588cd34e19388337a2ecb0a0cb3f796a37647e24

This patch follows the previous refactoring patch, and further reduces
the number of calls made to scan_row_mbmi and scan_col_mbmi by going
through partition blocks instead of mi blocks. This patch doesn't change
bitstream, which was proven by Borg test result.

The baseline decoder speed test on 1080p clip showed that the average
decoder speedup was 1.1%.(fps: 32.626 --> 32.994)

Change-Id: Ic375ae5d682c7454e2f2a2fcf8baa6b4b438d9a6

Change-Id: I620fab24173eef846510a74f4186c315f6623052

This allows gm parameters to be written when error resilient mode
is enabled but avoids depending on stored parameters from the
previous frame.

BUG=aomedia:741

Change-Id: Iae455a346e0eebb18e2bcfadc8826c75aedfee99

Instead of forward-declaring AV1_COMMON and MACROBLOCKD,
move the dependent struct and function prototype closer
to where they are used and after these types are defined.

Change-Id: I75f005b46ef322a6fcbc01377b8dded1637c5f73

Change-Id: I5d60cc74233bf7ddf56bcd0c3e8b8d012a6e5a63

The calls to count_segs appear very different from each other, but
this is really just because of line wrapping (enforced by the
clang-format linter). This patch produces equivalent code but the
results are far easier to read.

Change-Id: I9d548a4988d4ef23f71fbd8afc855eb8a43f7bd5

This function is given a bsize and an mi array and has to figure out
what partition to use to divide the given bsize in the direction of
the sizes it finds in the mi array. (Since each block size can be
reached by only one sequence of partitions, this can be done
unambiguously)

The previous version was correct, working by looking up entries in the
partition_lookup array. Unfortunately, that lookup isn't quite enough
when CONFIG_EXT_PARTITION_TYPES is true, so it then had to do some
slightly confusing computations to fix things up after the fact.

The new version should be more self-explanatory and doesn't work by
looking things up in a magic array. It looks up the width and height
corresponding to bsize and compares them with the width and height
corresponding to the sb_type at mi_row,mi_col in the mi array. When
CONFIG_EXT_PARTITION_TYPES is false, this is all you need, and the
four corresponding cases can be found by a lookup in an array of 4
elements.

With extended partition types and a sufficiently large block, you have
to do a bit more searching. For example, if bsize is BLOCK_16X16 and
the subsize is BLOCK_8X8, the partition might be PARTITION_SPLIT, but
it might be one of PARTITION_HORZ_A or PARTITION_VERT_A instead. The
new code adds some comments to explain what's going on.

A nice side-effect of rewriting get_partition in this way is that it
lets us completely dispense with the partition_lookup array.

The patch also fixes comments for the A/B extended partitions in
enums.h, which were slightly backwards (a "horizontal split means two
blocks vertically above one another)

Change-Id: I4b48189103aa63e1859f25a15d7690d53ca7baf5

By using a combination of a small qmlevel and large qindex,
it is possible for the quantizer 'dqv' to end up > 2^15.
This currently causes dqv to overflow, so widen its type
to avoid that.

Note that this is certainly an extreme case, but not entirely
useless - if using a 32x32 transform, the dequantized value
can be as small as dqv/2, which may still fit in 16 bits.

Change-Id: I0451e70ba6b17620d94386a98d873f0a8c615635

Using an assert (which was compiled out) meant we got to a
use-after-free in bug 729.

BUG=aomedia:729

Change-Id: I4d0fa5e80322c395f94565c76e2285782c328505

Turn off compound modes as long as one of the dimensions
is less than 8.

Imapct on AWCY (0.05% increase in BDRATE)
https://arewecompressedyet.com/?job=debargha-nocdef-sub8c8nc-0907%402017-09-07T20%3A28%3A38.251Z&job=debargha-nocdef-0907%402017-09-07T14%3A42%3A17.170Z

Change-Id: I4a70890c04149246a50e60990dede21cb8052fad

Remove mistakenly leftover code from commit
e30a47ca.

Note that this change force-disables activity masking in PVQ until we can add some
bitstream signaling for it.

Change-Id: I5279f82e7977733ebc40f5aed844ecec1279992a

Change-Id: Ie92a17d6c530bdd65464fbc599ce92561b8c0428

This won't change the coding statistics.

Change-Id: Ibe3790836c59462f830029528d88a6e615efa493

Before LR the border extension needs to be consistent.
Otherwise there will be inconsistency at the frame boundaries
causing mismatches.

BUG=aomedia:742
BUG=aomedia:744
BUG=aomedia:745

Change-Id: I5b0c51aa1493c58f7acc706bf686ce711eadd75d

This reverts commit 72c99e1a.

No metrics on this commit.

Change-Id: I9fc350b25e710c3d5d6d8299ab5348e8b31b39ea

Replaces the orthonormal Type-IV DST with an orthonormal Type-VII DST
 in od_bin_fdst4() and od_bin_idst4()

Change-Id: I4ff0888e740d8cc063a2e5deaeceef7cb0d80485

The code currently pads the decoded frame width and height to a
multiple of 8 luma pixels, but there is a TODO suggesting that
this may be changed to only require a multiple of 4 in future.

But, as per the comments on the linked bug, there are good reasons
to keep the decoded width and height as multiples of 8. So delete
the outdated TODO and instead outline the reasons why the current
behaviour is helpful.

BUG=aomedia:727

Change-Id: I2340bbcd740afe74c2e6fb3cf2e7a420db2b4f40