Commit Graph

15237 Commits

Author SHA1 Message Date
Geza Lore
5aa655f79a Reintroduce VAR_TX fix
Reintroduce part of
Iaf2b717e6b8626b2b6a03226127221b776b49884
Which was later reverted in
I4c5b40ec63a6f19521191d3c730af87db3c4bc00

Change-Id: If3e5610ba3985ae7b4d952d8e616982465ac667a
2016-01-26 15:27:47 +00:00
Geza Lore
f33755ef1f 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
2016-01-26 15:21:07 +00:00
Geza Lore
753dcbe5f9 Merge "Update VAR_TX related context when using SUPERTX." into nextgenv2 2016-01-26 15:20:32 +00:00
Yaowu Xu
f512a311f2 Merge branch 'master' into nextgenv2 2016-01-26 05:00:05 -08:00
Geza Lore
22cfc841d5 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
2016-01-26 01:18:56 +00:00
Geza Lore
56686b4514 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
2016-01-26 01:18:56 +00:00
Geza Lore
e7c0e157d2 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
2016-01-26 01:18:56 +00:00
Debargha Mukherjee
a8122bb957 A minor assert fix for supertx
Change-Id: I532fff64ccaa1f38240ba7ca5ce2f7e1eb531771
2016-01-25 17:15:51 -08:00
Debargha Mukherjee
3e63170fdd Merge "Some supertx fixes" into nextgenv2 2016-01-25 22:21:53 +00:00
James Zern
3a2ad10de2 Merge "Code clean of sad4xNx4D_sse" 2016-01-25 20:57:15 +00:00
Debargha Mukherjee
9a8a6a1b35 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
2016-01-25 10:36:41 -08:00
Zoe Liu
9e612763f0 Cleaned a code in define_gf_group() for firstpass
Change-Id: I0b8eff9e74e3446024482d011ec4fb28b19136d3
2016-01-25 09:38:45 -08:00
Yaowu Xu
33768c24af Merge branch 'master' into nextgenv2 2016-01-24 05:00:05 -08:00
Alex Converse
27c13712c9 Merge "Short circuit flat blocks when coding screen content at realtime speed." 2016-01-23 02:04:37 +00:00
Scott LaVarnway
be916257bd Merge "VP9: Do not set up/left available in vp9_first_pass()" 2016-01-22 21:53:26 +00:00
Yue Chen
968bbc7bb2 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
2016-01-22 13:52:16 -08:00
Alex Converse
7da6324cab 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
2016-01-22 12:40:45 -08:00
Scott LaVarnway
78fadcd063 VP9: Do not set up/left available in vp9_first_pass()
These values are set in set_mi_row_col()

Change-Id: Ie44c9d2dcb79797a8473b5fc29394277a7117547
2016-01-22 11:30:51 -08:00
Yaowu Xu
2ab58d812f Fix a minor typo
a wrong variable.

Change-Id: I9fa02621dbdc508a12ba1ef9c8f98dbeab68e958
2016-01-22 10:57:16 -08:00
Yaowu Xu
53c9ffd06f Fix a merge mistake
Change-Id: I0769ca1abd42521a24078dc9ba7093ccb5e362ee
2016-01-22 10:48:10 -08:00
Debargha Mukherjee
11d3899c60 Merge "A fix for a missing tx_type" into nextgenv2 2016-01-22 18:36:46 +00:00
Yaowu Xu
307c84c352 Merge "Properly set segmentation for error-resilient mode" 2016-01-22 16:52:41 +00:00
Yaowu Xu
94322a9f9b Merge branch 'master' into nextgenv2
Change-Id: I0a82fa1cbe4ee7c7831d2d174f140a40d09a06c5
2016-01-22 08:46:06 -08:00
Debargha Mukherjee
4e406f702c A fix for a missing tx_type
Change-Id: I165cd06256175edb7739020379ba3098251f4a7c
2016-01-22 03:29:39 -08:00
Debargha Mukherjee
ade60d4fe9 Merge "Fixes for var_tx when ext_tx is not enabled" into nextgenv2 2016-01-22 06:09:36 +00:00
Debargha Mukherjee
0a5c176011 Merge "Some fixes on tx size/type selection" into nextgenv2 2016-01-22 04:02:11 +00:00
Yue Chen
91f55290ef Merge "Relocate supertx prob update" into nextgenv2 2016-01-22 03:45:38 +00:00
Julia Robson
9fe188e4a5 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
2016-01-21 17:46:03 -08:00
Debargha Mukherjee
8d69a6e816 Some fixes on tx size/type selection
For ext_tx experiment.

Change-Id: Ie37b9b456b09bde8b606fb978fee4cca8d0326b7
2016-01-21 17:45:30 -08:00
Debargha Mukherjee
b1d49f24b4 Build fix for vp9
Change-Id: I1b93ea826487f9033de70fc4563ade410fe07d74
2016-01-21 17:41:36 -08:00
Yue Chen
fb29aec42c 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
2016-01-21 17:41:21 -08:00
Marco Paniconi
732242ab3d Merge "vp9 pickmode: Fix the shift on negative value." 2016-01-22 00:09:27 +00:00
Marco
dbc963a42f 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
2016-01-21 15:16:52 -08:00
Alex Converse
a61ac8e2e7 Merge "Remove the vestigal skip_eob feature from tokenization." 2016-01-21 23:11:14 +00:00
James Zern
604e975f93 Merge "vp9_aq_cyclicrefresh: fix double->float warning" 2016-01-21 23:08:02 +00:00
James Zern
bd4501fe28 vp9_aq_cyclicrefresh: fix double->float warning
Change-Id: I39d3a150de9a77967a4e25ca89d2eda51fd18d47
2016-01-21 14:07:33 -08:00
Yaowu Xu
8c4360ed89 Merge "Fix a typo in ctrl_set_render_size()" 2016-01-21 20:06:29 +00:00
Yaowu Xu
b6cefbf052 Fix a typo in ctrl_set_render_size()
Change-Id: I6a7fa6d0ae8ce8de7b426189d5eb00413f21bc17
2016-01-21 11:19:08 -08:00
Alex Converse
4326cffa65 Merge "Tie the bit cost scale to a define." 2016-01-21 19:17:56 +00:00
Debargha Mukherjee
09f3615606 Merge "Loop restoration filter" into nextgenv2 2016-01-21 17:43:39 +00:00
Debargha Mukherjee
0880b42129 Merge "Making the forward transform consistent with high bit depth" into nextgenv2 2016-01-21 15:22:48 +00:00
Marco
a925173da0 vp9-aqmode=3: Reduce condition below which we turn off delta-qp.
Add TODO to consider turning-off more smoothly.

Change-Id: Id6dd4c5c515f5b09c388af965eb27dbe24924362
2016-01-20 20:26:40 -08:00
hui su
19de4d9408 Fix build with --enable-internal-stats
Change-Id: I9649647cc5755c8411ddd56f92e7f96d421c2663
2016-01-20 18:02:30 -08:00
Debargha Mukherjee
84ca7a9f0f 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
2016-01-20 17:59:46 -08:00
Marco Paniconi
4ec5d7dd18 Merge "vp9-svc: Allow frame-dropping in svc datarate control tests." 2016-01-21 01:39:53 +00:00
Alex Converse
5bd99f83d9 Remove the vestigal skip_eob feature from tokenization.
Repack TOKENEXTRA fields.

Speed impact within measurment margin of error.

Change-Id: I9a6d1dde1bb4a0766b02d0cb74c871ddde907cde
2016-01-20 17:13:59 -08:00
Marco
f5c07b6803 vp9-svc: Allow frame-dropping in svc datarate control tests.
Remove comment(s) and enable frame-dropper for tests.
Frame dropper for 1 pass svc was fixed a while ago:
https://chromium-review.googlesource.com/#/c/309230/

Change-Id: I5fd3192825b22e562db9210d3dc7b246a1799d8d
2016-01-20 14:00:42 -08:00
Marco
4800e179fe vp8-denoiser: Fix threshold for skin map.
Make it consistent with the comment/intended behavior,
that is, only denoise if current block is zero_mv.

Change-Id: I3909761e802e80089752a493ab3646dc32698ded
2016-01-20 13:48:46 -08:00
Yaowu Xu
31a94868b5 Merge "Remove duplicate definitions" 2016-01-20 18:12:51 +00:00
Yaowu Xu
a40d486215 Merge branch 'master' into nextgenv2
Change-Id: I4e5dd38caa9608252235265c7f2342b183f99815
2016-01-20 09:24:41 -08:00