]>
granicus.if.org Git - libvpx/log
Angie Chiang [Tue, 2 Feb 2016 02:32:16 +0000 (02:32 +0000)]
Merge "Pass filter type instead of filter array" into nextgenv2
Angie Chiang [Mon, 1 Feb 2016 20:49:15 +0000 (12:49 -0800)]
Pass filter type instead of filter array
Change-Id: I25f2149ddaa332722f7ab82e8f832a253c4b6ab3
Debargha Mukherjee [Tue, 2 Feb 2016 00:17:14 +0000 (00:17 +0000)]
Merge "Refactor to separate restoration from loop filter" into nextgenv2
Yaowu Xu [Mon, 1 Feb 2016 20:29:36 +0000 (12:29 -0800)]
Fix automerge errors
Change-Id: I24d415bafe617eac894427088d7b2fbe0b7e04d7
Yaowu Xu [Mon, 1 Feb 2016 13:00:05 +0000 (05:00 -0800)]
Merge branch 'master' into nextgenv2
Scott LaVarnway [Mon, 1 Feb 2016 12:52:15 +0000 (12:52 +0000)]
Merge "VP9: Refactor read_inter_block_mode_info()"
Yaowu Xu [Sun, 31 Jan 2016 13:00:05 +0000 (05:00 -0800)]
Merge branch 'master' into nextgenv2
Yaowu Xu [Sun, 31 Jan 2016 04:38:39 +0000 (04:38 +0000)]
Merge "Enable sse2 version of inverse wht for hbd build"
Yaowu Xu [Sat, 30 Jan 2016 13:00:05 +0000 (05:00 -0800)]
Merge branch 'master' into nextgenv2
James Zern [Sat, 30 Jan 2016 00:27:45 +0000 (00:27 +0000)]
Merge changes I51d42a50,Id8aaf820,I6af91959
* changes:
configure: extend armv7 hf target autodetect
configure: remove default CROSS for arm targets
configure: avoid default when CROSS is set to null
James Zern [Sat, 30 Jan 2016 00:17:16 +0000 (00:17 +0000)]
Merge changes If13946e4,I61a1814d,I2ca9aa3c,I44d91eaa
* changes:
intrapred: protect functions w/CONFIG check
vp9_noise_estimate: protect copy_frame w/CONFIG check
vp8_cx_iface: delete 3 unused functions
vp8: mark intra_prediction_down_copy inline
Hui Su [Fri, 29 Jan 2016 23:58:49 +0000 (23:58 +0000)]
Merge "Refactor prob_diff_update_savings_search_model"
Debargha Mukherjee [Fri, 29 Jan 2016 23:30:19 +0000 (15:30 -0800)]
Refactor to separate restoration from loop filter
Change-Id: Iab517862d957f3aa2a664e9349d57bbf424febb3
Scott LaVarnway [Fri, 29 Jan 2016 23:30:27 +0000 (15:30 -0800)]
VP9: Refactor read_inter_block_mode_info()
Change-Id: I8e6a1a1b6eec17d0076660b559f6fc120a78c538
Yaowu Xu [Fri, 29 Jan 2016 22:47:56 +0000 (14:47 -0800)]
Enable sse2 version of inverse wht for hbd build
Change-Id: If8f5efd701a11c8a7ad3078d10ec3cd0fe27667e
Debargha Mukherjee [Fri, 29 Jan 2016 20:57:04 +0000 (20:57 +0000)]
Merge "Cosmetic changes to loop restoration" into nextgenv2
hui su [Fri, 29 Jan 2016 20:49:19 +0000 (12:49 -0800)]
Refactor prob_diff_update_savings_search_model
Change-Id: Icfab58e880285d5af00985dc20db2174a67d1fe8
Yaowu Xu [Fri, 29 Jan 2016 00:29:29 +0000 (16:29 -0800)]
SSSE3 idct8x8 functions for highbitdpeth build
This commit changes SSSE3 optimized idct8x8 functions to work with
highbitdepth build.
With this commit and the previous one that enabled SSSE3 idct32x32
functions, tests showed virtually no difference on decoding speed for
file fdJc1_IBKJA.248.webm for the build with -enable-vp9-highbitdpeth
option and the build without the option.
Change-Id: Ibe0634149ec70e8b921e6b30171664b8690a9c45
Marco Paniconi [Fri, 29 Jan 2016 18:55:13 +0000 (18:55 +0000)]
Merge "Fix failure with libvpx__unit_tests-multi-target"
Marco Paniconi [Fri, 29 Jan 2016 18:43:27 +0000 (18:43 +0000)]
Merge "vp9-denoiser: Force increase_denoising = 0 for skin block."
Marco [Fri, 29 Jan 2016 18:00:21 +0000 (10:00 -0800)]
Fix failure with libvpx__unit_tests-multi-target
Introduced in: https://chromium-review.googlesource.com/#/c/323215/
For now exclude the computation of vp_compute_skin_block for highbitdepth.
Change-Id: I18107a862c17ba869438ec3616a787bf173736c4
Marco [Fri, 29 Jan 2016 16:36:03 +0000 (08:36 -0800)]
vp9-denoiser: Force increase_denoising = 0 for skin block.
Change-Id: Ic5af41db58a3daad4103175b4b4b9300675020c0
Zoe Liu [Fri, 29 Jan 2016 16:31:41 +0000 (16:31 +0000)]
Merge "Cleaned code in firstpass related to altref frame"
Yaowu Xu [Wed, 27 Jan 2016 23:25:42 +0000 (15:25 -0800)]
Enable hbd_build to use SSSE3optimized functions
This commit changes the SSSE3 assembly functions for idct32x32 to
support highbitdepth build.
On test clip fdJc1_IBKJA.248.webm, this cuts the speed difference
between hbd and lbd build from between 3-4% to 1-2%.
Change-Id: Ic3390e0113bc1ca5bba8ec80d1795ad31b484fca
Yaowu Xu [Fri, 29 Jan 2016 01:28:43 +0000 (01:28 +0000)]
Merge "Add a vp9 test vector"
Marco Paniconi [Fri, 29 Jan 2016 01:19:30 +0000 (01:19 +0000)]
Merge "vp9: encoder-breakout: Modify UV thresholds."
Zoe Liu [Fri, 29 Jan 2016 00:36:22 +0000 (16:36 -0800)]
Cleaned code in firstpass related to altref frame
Change-Id: I6dbddf500fd4f64216774685700efea972165470
Yaowu Xu [Thu, 28 Jan 2016 16:50:50 +0000 (08:50 -0800)]
Add a vp9 test vector
This test vector exercises various idct32x32 functions in decoder.
Change-Id: Ic9453f50ef4723bdaace9bd80f32540d061ef2a1
James Zern [Thu, 28 Jan 2016 20:53:41 +0000 (12:53 -0800)]
vp9/10: fix encoder crash on flush
the lookahead buffer allocation is deferred to receipt of the first
frame to allow profile changes. if the encoder was flushed before
supplying any frames the encoder would crash trying to dereference the
NULL buffer. vp8 is unaffected.
fixes mozilla bug:
https://bugzilla.mozilla.org/show_bug.cgi?id=
1237848
Change-Id: Icee4b64de760476eee0d33b568f0a1010335ff13
Marco [Thu, 28 Jan 2016 20:10:38 +0000 (12:10 -0800)]
vp9: encoder-breakout: Modify UV thresholds.
Set UV thresholds to 0 for encoder breakout if
block is detected as skin.
Change-Id: Iaaa30b5cc8bae4cf611f8ea8eba693b730e32a3f
JackyChen [Tue, 26 Jan 2016 10:01:10 +0000 (18:01 +0800)]
VPX skin map improvement.
Use multiple clusters instead of one and decrease
the distance thresholds.
Add a define to switch between models.
Default is set to existing (1 cluster) model.
Change-Id: I802cd9bb565437ae8983ef39453939f5d5073bb1
Marco Paniconi [Thu, 28 Jan 2016 17:01:19 +0000 (17:01 +0000)]
Merge "vp9 non-rd mode: Modification for detected skin areas."
Yaowu Xu [Thu, 28 Jan 2016 15:32:05 +0000 (07:32 -0800)]
Merge branch 'master' into nextgenv2
Scott LaVarnway [Thu, 28 Jan 2016 12:31:22 +0000 (12:31 +0000)]
Merge "VP9: Refactor dec_build_inter_predictors_sb()"
Marco [Fri, 22 Jan 2016 19:45:31 +0000 (11:45 -0800)]
vp9 non-rd mode: Modification for detected skin areas.
If a superblock contains alot of "skin" then force split
of 64x64 partition, and make some adjustments in mode selection.
This helps to reduce artifacts on moving face/skin areas at low bitrates.
Little/no change in metrics: avgPSNR/SSIM down by ~0.12%.
Small encoding time increase < 1%.
Change-Id: Ic57f52148c3716f391419fab0530d916e4c1d186
Debargha Mukherjee [Wed, 27 Jan 2016 21:01:27 +0000 (13:01 -0800)]
Cosmetic changes to loop restoration
Also adds a normalized filtering function to be used later.
Change-Id: I30e2140e664db635602f26a73b81ce8e008dff5e
Hui Su [Thu, 28 Jan 2016 01:04:50 +0000 (01:04 +0000)]
Merge "Fix some typos."
hui su [Mon, 25 Jan 2016 18:19:16 +0000 (10:19 -0800)]
Fix some typos.
Change-Id: I32aacd014df6c927cf2893dc096cbe6ec7604b9b
Marco Paniconi [Wed, 27 Jan 2016 21:38:35 +0000 (21:38 +0000)]
Merge "vp9-aqmode=3: Limit on period for golden update."
Alex Converse [Wed, 27 Jan 2016 20:54:07 +0000 (20:54 +0000)]
Merge "Switch to 9-bit rate cost constants built on a 256 probability denominator."
Marco [Wed, 27 Jan 2016 20:01:58 +0000 (12:01 -0800)]
vp9-aqmode=3: Limit on period for golden update.
For aqmode=3, golden period update is set based on period of cyclic refresh.
Put a limit on max golden period update, for now set to 40.
And fix comment.
Change-Id: Icb61dd87c796cce2a5f5f7331c6a129540994696
Alex Converse [Wed, 13 Jan 2016 22:21:30 +0000 (14:21 -0800)]
Switch to 9-bit rate cost constants built on a 256 probability denominator.
-.220 BDRATE derf: https://x20web.corp.google.com/~aconverse/results/cost256_derf.html
-.675 BDRATE hevcmr: https://x20web.corp.google.com/~aconverse/results/cost256_hevcmr.html
Change-Id: Ifb1646d8ce65ffe0eff9953a911b1b88735b335f
Debargha Mukherjee [Wed, 27 Jan 2016 19:16:55 +0000 (19:16 +0000)]
Merge "Fixes ext-interp experiment" into nextgenv2
Debargha Mukherjee [Wed, 27 Jan 2016 16:09:39 +0000 (08:09 -0800)]
Fixes ext-interp experiment
Fixes integer pel MV usage for the sub8x8 case, which fixes a
rare mismatch issue.
Also adds some other minor missing code related to filter threshes.
Change-Id: I6b07e6cf9b287ba4b5bd6599af4a7412e50b3bdc
Scott LaVarnway [Wed, 27 Jan 2016 15:37:59 +0000 (07:37 -0800)]
VP9: Refactor dec_build_inter_predictors_sb()
Change-Id: I336fe2006288214977b5ee854124ee7b5bba5796
Yaowu Xu [Wed, 27 Jan 2016 13:00:06 +0000 (05:00 -0800)]
Merge branch 'master' into nextgenv2
James Zern [Wed, 27 Jan 2016 04:13:57 +0000 (20:13 -0800)]
intrapred: protect functions w/CONFIG check
d207e, d63e, d45e are only used with CONFIG_MISC_FIXES
Change-Id: If13946e483c4d0ccaa3e1d60dc14216c06d5a219
James Zern [Wed, 27 Jan 2016 04:09:36 +0000 (20:09 -0800)]
vp9_noise_estimate: protect copy_frame w/CONFIG check
This is only used with CONFIG_VP9_TEMPORAL_DENOISING
Change-Id: I61a1814d43d71235077a2426f7bfce141677a79a
James Zern [Wed, 27 Jan 2016 04:08:37 +0000 (20:08 -0800)]
vp8_cx_iface: delete 3 unused functions
vp8e_update_entropy, vp8e_update_reference, vp8e_use_reference have been
unused since:
a046f56 vp8cx: remove deprecated reference/entropy controls
Change-Id: I2ca9aa3c3f1a209a534a1492432b16852e3d8629
James Zern [Wed, 27 Jan 2016 04:07:42 +0000 (20:07 -0800)]
vp8: mark intra_prediction_down_copy inline
avoids -Wunused-function warnings when INLINE is set
Change-Id: I44d91eaa7efba7bc2427501fb9f63a93f32aaa7f
Yunqing Wang [Tue, 26 Jan 2016 19:00:06 +0000 (19:00 +0000)]
Merge "Fix a clang/win build error"
Yunqing Wang [Tue, 26 Jan 2016 18:10:07 +0000 (10:10 -0800)]
Fix a clang/win build error
Made the definition of THREAD_FUNCTION consistent.
Change-Id: I1ac099484e201e359298ed16de0b81ec781075ce
Alex Converse [Tue, 26 Jan 2016 17:59:59 +0000 (17:59 +0000)]
Merge "vp9: Be more responsive to massive overshoot."
Geza Lore [Tue, 26 Jan 2016 15:27:43 +0000 (15:27 +0000)]
Reintroduce VAR_TX fix
Reintroduce part of
Iaf2b717e6b8626b2b6a03226127221b776b49884
Which was later reverted in
I4c5b40ec63a6f19521191d3c730af87db3c4bc00
Change-Id: If3e5610ba3985ae7b4d952d8e616982465ac667a
Geza Lore [Tue, 26 Jan 2016 14:40:32 +0000 (14:40 +0000)]
Always recode SUPERTX blocks.
There is still an assertion failure when tokenizing transform
coefficients for a supertx block, due to the eob not being set
consistently with the coefficients, so we always recode supertx blocks
for now. Also added further PICK_MODE_CONTEXT instances to avoid
potential clash between horizontal/vertical/split partition SUPERTX
trials.
Change-Id: I5f3da1fa0d8d20fc21face170487e1a285fd1cc6
Geza Lore [Tue, 26 Jan 2016 15:20:32 +0000 (15:20 +0000)]
Merge "Update VAR_TX related context when using SUPERTX." into nextgenv2
Yaowu Xu [Tue, 26 Jan 2016 13:00:05 +0000 (05:00 -0800)]
Merge branch 'master' into nextgenv2
Geza Lore [Tue, 19 Jan 2016 18:15:40 +0000 (18:15 +0000)]
Update VAR_TX related context when using SUPERTX.
The encoder did not update left_txfm_context and above_txfm_context in
MACROBLOCKD (used for choosing the probability context for the vartx
split bits) when the supertx bit was set for a block. The deoder on the
other hand did update these for supertx blocks. The encoder used these
to compute the context counts, which the packer then uses to adapt it's
probabilities. This results in the packer and the decoder using
different probabilities.
This patch harmonizes the encoder and the decoder by making the encoder
update the mentioned context for supertx coded blocks.
Change-Id: I3a22132124b1bce2ee501d640ceab374b19e3ca1
Geza Lore [Mon, 25 Jan 2016 15:46:59 +0000 (15:46 +0000)]
Initalize mbmi->tx_size during rdopt.
This is necessary when using SUPERTX, as the bitstream packer relies on
tx_size being set correctly to decide whether to output the block using
supertx or not.
Change-Id: I79e776b3b810f4a15b9dbc6afdd6fc90c73c8934
Geza Lore [Thu, 21 Jan 2016 10:46:33 +0000 (10:46 +0000)]
Set inter_tx_size for supertx coded blocks.
The loop filter relies on inter_tx_size in MB_MODE_INFO being set
properly when VAR_TX is enabled. Supertx coded blocks did not set this
previously at all, and the differing garbage values eventually resulted
in in a YUV mismatch between encoder and decoder after loop filtering.
This patch fixes this by setting inter_tx_size to the proper supertx
size in both the encoder and the decoder. This should also mean that
loop filtering is done at the proper transform boundaries, even when
supertx or vartx is being used.
Change-Id: I41a564cd6d34ce4a8313ad4efa89d905f5ead731
Debargha Mukherjee [Tue, 26 Jan 2016 01:13:26 +0000 (17:13 -0800)]
A minor assert fix for supertx
Change-Id: I532fff64ccaa1f38240ba7ca5ce2f7e1eb531771
Debargha Mukherjee [Mon, 25 Jan 2016 22:21:53 +0000 (22:21 +0000)]
Merge "Some supertx fixes" into nextgenv2
James Zern [Mon, 25 Jan 2016 20:57:15 +0000 (20:57 +0000)]
Merge "Code clean of sad4xNx4D_sse"
Alex Converse [Mon, 25 Jan 2016 20:03:50 +0000 (12:03 -0800)]
vp9: Be more responsive to massive overshoot.
Limit oscilation detection in the case where overshoot is very very
large.
This keeps the 9-bit cost patch from breaking the DownUp reisze test.
The patch pushed us to an 11% undershoot right before a scene cut
causing a 1200% overshoot. (Whereas before we were undershooting by
only 6% before overshooting by 1200%).
Change-Id: Id90ccfab8aba872ccadc45b73b3bb097b895677f
Debargha Mukherjee [Fri, 22 Jan 2016 22:52:38 +0000 (14:52 -0800)]
Some supertx fixes
Fixes some of the issues introduced by a merge from master.
derflr: -0.893% BDRATE
hevcmr: -1.667% BDRATE
Change-Id: I4c5b40ec63a6f19521191d3c730af87db3c4bc00
Zoe Liu [Sat, 23 Jan 2016 01:13:03 +0000 (17:13 -0800)]
Cleaned a code in define_gf_group() for firstpass
Change-Id: I0b8eff9e74e3446024482d011ec4fb28b19136d3
Yaowu Xu [Sun, 24 Jan 2016 13:00:05 +0000 (05:00 -0800)]
Merge branch 'master' into nextgenv2
Alex Converse [Sat, 23 Jan 2016 02:04:37 +0000 (02:04 +0000)]
Merge "Short circuit flat blocks when coding screen content at realtime speed."
Scott LaVarnway [Fri, 22 Jan 2016 21:53:26 +0000 (21:53 +0000)]
Merge "VP9: Do not set up/left available in vp9_first_pass()"
Yue Chen [Wed, 20 Jan 2016 00:45:45 +0000 (16:45 -0800)]
Adding new compound modes to EXT_INTER experiment
Combinations of different mv modes for two reference frames
are allowed in compound inter modes. 9 options are enabled,
including NEAREST_NEARESTMV, NEAREST_NEARMV, NEAR_NEARESTMV,
NEAREST_NEWMV, NEW_NEARESTMV, NEAR_NEWMV, NEW_NEARMV, ZERO_ZEROMV,
and NEW_NEWMV.
This experiment is mostly deported from the nextgen branch.
It is made compatible with other experiments
Coding gain of EXT_INTER(derflr/hevcmr/hevchd): 0.533%/0.728%/0.639%
Change-Id: Id47e97284e6481b186870afbad33204b7a33dbb0
Alex Converse [Wed, 20 Jan 2016 22:28:09 +0000 (14:28 -0800)]
Short circuit flat blocks when coding screen content at realtime speed.
In inter mode search skip all modes except NEARESTMV and DC_PRED.
10% less encode latency for large frames using the chromium remoting_perftests.
+0.313% BDRATE on the screencast set at speed -6.
Change-Id: Ib97a39dd8bcdeab545509e0e02d78ce7033f8c63
Scott LaVarnway [Fri, 22 Jan 2016 17:22:53 +0000 (09:22 -0800)]
VP9: Do not set up/left available in vp9_first_pass()
These values are set in set_mi_row_col()
Change-Id: Ie44c9d2dcb79797a8473b5fc29394277a7117547
Yaowu Xu [Fri, 22 Jan 2016 18:57:16 +0000 (10:57 -0800)]
Fix a minor typo
a wrong variable.
Change-Id: I9fa02621dbdc508a12ba1ef9c8f98dbeab68e958
Yaowu Xu [Fri, 22 Jan 2016 18:48:10 +0000 (10:48 -0800)]
Fix a merge mistake
Change-Id: I0769ca1abd42521a24078dc9ba7093ccb5e362ee
Debargha Mukherjee [Fri, 22 Jan 2016 18:36:46 +0000 (18:36 +0000)]
Merge "A fix for a missing tx_type" into nextgenv2
Yaowu Xu [Fri, 22 Jan 2016 16:52:41 +0000 (16:52 +0000)]
Merge "Properly set segmentation for error-resilient mode"
Yaowu Xu [Fri, 22 Jan 2016 16:45:37 +0000 (08:45 -0800)]
Merge branch 'master' into nextgenv2
Change-Id: I0a82fa1cbe4ee7c7831d2d174f140a40d09a06c5
Debargha Mukherjee [Fri, 22 Jan 2016 11:29:39 +0000 (03:29 -0800)]
A fix for a missing tx_type
Change-Id: I165cd06256175edb7739020379ba3098251f4a7c
Debargha Mukherjee [Fri, 22 Jan 2016 06:09:36 +0000 (06:09 +0000)]
Merge "Fixes for var_tx when ext_tx is not enabled" into nextgenv2
Debargha Mukherjee [Fri, 22 Jan 2016 04:02:11 +0000 (04:02 +0000)]
Merge "Some fixes on tx size/type selection" into nextgenv2
Yue Chen [Fri, 22 Jan 2016 03:45:38 +0000 (03:45 +0000)]
Merge "Relocate supertx prob update" into nextgenv2
Julia Robson [Thu, 21 Jan 2016 17:14:29 +0000 (17:14 +0000)]
Fixes for var_tx when ext_tx is not enabled
This patch fixes a couple of issues caused by change-id:
I15d20ce5292b70f0c2b4ba55c1f1318181481596
Changes to the code for when the ext_tx experiment is not enabled
were merged from master but as var_tx does not exist on master
the changes had not been applied to the case when var_tx experiment
is enabled
Change-Id: Iaf2b717e6b8626b2b6a03226127221b776b49884
Debargha Mukherjee [Wed, 6 Jan 2016 22:36:13 +0000 (14:36 -0800)]
Some fixes on tx size/type selection
For ext_tx experiment.
Change-Id: Ie37b9b456b09bde8b606fb978fee4cca8d0326b7
Debargha Mukherjee [Fri, 22 Jan 2016 01:41:36 +0000 (17:41 -0800)]
Build fix for vp9
Change-Id: I1b93ea826487f9033de70fc4563ade410fe07d74
Yue Chen [Fri, 22 Jan 2016 01:08:52 +0000 (17:08 -0800)]
Relocate supertx prob update
Move it from vp10_adapt_intra_frame_probs() to
vp10_adapt_inter_frame_probs() because intra frames do not use
supertx.
Change-Id: I28c7391944848666054d4b990ac17a8ae08aaaee
Marco Paniconi [Fri, 22 Jan 2016 00:09:27 +0000 (00:09 +0000)]
Merge "vp9 pickmode: Fix the shift on negative value."
Marco [Thu, 21 Jan 2016 23:13:10 +0000 (15:13 -0800)]
vp9 pickmode: Fix the shift on negative value.
Error only affected svc 1 pass mode, introduced in:
https://chromium-review.googlesource.com/#/c/319609/
Change-Id: Ifc078c82d38b742966a81baaf5ad532f7dbc13a4
Alex Converse [Thu, 21 Jan 2016 23:11:14 +0000 (23:11 +0000)]
Merge "Remove the vestigal skip_eob feature from tokenization."
James Zern [Thu, 21 Jan 2016 23:08:02 +0000 (23:08 +0000)]
Merge "vp9_aq_cyclicrefresh: fix double->float warning"
James Zern [Thu, 21 Jan 2016 22:07:33 +0000 (14:07 -0800)]
vp9_aq_cyclicrefresh: fix double->float warning
Change-Id: I39d3a150de9a77967a4e25ca89d2eda51fd18d47
Yaowu Xu [Thu, 21 Jan 2016 20:06:29 +0000 (20:06 +0000)]
Merge "Fix a typo in ctrl_set_render_size()"
Yaowu Xu [Thu, 21 Jan 2016 19:19:08 +0000 (11:19 -0800)]
Fix a typo in ctrl_set_render_size()
Change-Id: I6a7fa6d0ae8ce8de7b426189d5eb00413f21bc17
Alex Converse [Thu, 21 Jan 2016 19:17:56 +0000 (19:17 +0000)]
Merge "Tie the bit cost scale to a define."
Debargha Mukherjee [Thu, 21 Jan 2016 17:43:39 +0000 (17:43 +0000)]
Merge "Loop restoration filter" into nextgenv2
Debargha Mukherjee [Thu, 21 Jan 2016 15:22:48 +0000 (15:22 +0000)]
Merge "Making the forward transform consistent with high bit depth" into nextgenv2
Marco [Thu, 21 Jan 2016 01:44:02 +0000 (17:44 -0800)]
vp9-aqmode=3: Reduce condition below which we turn off delta-qp.
Add TODO to consider turning-off more smoothly.
Change-Id: Id6dd4c5c515f5b09c388af965eb27dbe24924362
hui su [Thu, 21 Jan 2016 02:02:30 +0000 (18:02 -0800)]
Fix build with --enable-internal-stats
Change-Id: I9649647cc5755c8411ddd56f92e7f96d421c2663
Debargha Mukherjee [Tue, 19 Jan 2016 21:01:01 +0000 (13:01 -0800)]
Loop restoration filter
Current implementation is a bilateral filter whose
parameters are transmitted in the bitstream.
derflr: -0.647% BDRATE
hevcmr: -0.794% BDRATE
This is a prelimary patch. Various other variations are to
be investigated next, that will hopefully be less expensive
on the decoder side.
Change-Id: I50634ae8f5014ad0bf7432306348908a349d81e1
Marco Paniconi [Thu, 21 Jan 2016 01:39:53 +0000 (01:39 +0000)]
Merge "vp9-svc: Allow frame-dropping in svc datarate control tests."
Alex Converse [Wed, 20 Jan 2016 22:09:45 +0000 (14:09 -0800)]
Remove the vestigal skip_eob feature from tokenization.
Repack TOKENEXTRA fields.
Speed impact within measurment margin of error.
Change-Id: I9a6d1dde1bb4a0766b02d0cb74c871ddde907cde