17183 Commits

Author SHA1 Message Date
Scott LaVarnway
0840c6fb8f BUG FIX: undefined reference to `vp9_scale_and_extend_frame_c'
See https://bugs.chromium.org/p/webm/issues/detail?id=1145

Change-Id: I778ee07dc39a524e3f729bef47a7abeed51e0cee
2016-02-08 13:42:56 -08:00
Marco
2a3f994723 vp9-denoiser: Allow for checking skin for bsize <=32x32.
Change-Id: I6c99ac6fd3735e24cf668479c55bff313c0b4970
2016-02-08 13:01:38 -08:00
Jingning Han
f032c7eaed Merge "Account for sub8x8 block skip mode cost in RD decision" 2016-02-08 19:40:01 +00:00
Yaowu Xu
090eaadf20 Change to use local variables consistently
This commit does not change the computation, nor results.

Change-Id: I1a7bb47050220d970f075458b507c5e55d93b22e
2016-02-08 11:38:04 -08:00
Jingning Han
28e0393f23 Enable dynamic ref motion vector mode for compound inter block
This commit enables the dynamic reference motion vector coding mode
for the compound inter blocks.

Change-Id: Ibe78fd8de6989db392cd67a9d81a69d680345ba1
2016-02-08 11:24:15 -08:00
Yaowu Xu
204e77e059 Remove a flavor of SSIM that is never really used.
Change-Id: I61ea7f63acbcfeecd3f7dba5a5a38b980efc802b
2016-02-08 11:22:08 -08:00
Yaowu Xu
efe1b1dbf7 Set a max dB value for PSNR_HVS and FAST_SSIM
Now set at 100.0 instead of infinite

Change-Id: I41bae0c4bd95a26f9819584e7311b7945df1271a
2016-02-08 10:55:25 -08:00
Angie Chiang
355e586f45 Merge "Experiment: use 12 taps for sharp filter" into nextgenv2 2016-02-08 18:33:51 +00:00
Angie Chiang
eb71ef9235 Merge "add convolution function with adjustable length" into nextgenv2 2016-02-08 18:32:38 +00:00
Jingning Han
203bdd20fb Account for sub8x8 block skip mode cost in RD decision
Make this consistent with regular block size rate-distortion
optimization. It improves the compression performance:
derf    0.055%
hevcmr  0.129%

Change-Id: I112fe734f592c21bc7aa6efb7e3f269c4214ee7b
2016-02-08 10:18:51 -08:00
Yaowu Xu
c89bcae560 Merge "Fixed warnings of unused functions" into nextgenv2 2016-02-08 18:18:47 +00:00
Jingning Han
ac6d40ece8 Clean up in vp9_rd_pick_inter_mode_sb
Use local variable.

Change-Id: I0d3df36cf4536958a0cda422f6c30da50f0e0bbf
2016-02-08 10:15:02 -08:00
Yaowu Xu
3c28b4a8ff Fix msvc compiler warnings
There were a number of compiler warnings:
1. int16_t to uint8_t in recon_intra.c;
2. double to float conversions in psnrhvs.c
3. intptr_t to int in quantize.c
4. size_t to int32_t in decoder.c

Change-Id: Id95423b17779dcfa6cf39d9a90fe8cb8b910f5df
2016-02-08 10:14:08 -08:00
Yaowu Xu
d2ef991f9e Merge "Fix bad merge artifacts" into nextgenv2 2016-02-08 17:25:54 +00:00
Yaowu Xu
ac898d221f Normalize fdct8x8 in psnrhvs computation
This is to match the scale to the fdct8x8 used in original daala
psnrhvs computation.

Change-Id: Ic30b50747ba9c340bcb679f7439640046c69f90a
2016-02-08 17:13:18 +00:00
James Zern
ba7ebbb35c vp8: fix build with mingw+pthreads
Change-Id: Icc34a00759c95b7b8ac356cdcc4adae848b61431
2016-02-05 21:08:45 -08:00
Angie Chiang
b9d3fbe0c5 Experiment: use 12 taps for sharp filter
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
2016-02-05 17:39:37 -08:00
Angie Chiang
d5349112e8 add convolution function with adjustable length
Change-Id: I1a5b1e15a188ef11594d0c6ac0dbd42aac59cfca
2016-02-05 17:33:19 -08:00
Marco
ae98a61e22 vp9-dynamic resize: On a resize trigger frame refresh alt_ref.
For 1 pass real-time mode. No change in behavior as only last
and golden are used as references in 1 pass real-time mode.

Change-Id: Ie4655014eee1a8b271542f29d74b2c6f7fed54c9
2016-02-05 16:32:49 -08:00
Yaowu Xu
4d90ae4b49 Fixed warnings of unused functions
And enabled the warning flag in configure for vp10.

Change-Id: If556d6fac65755af3d6ed7fe71b8eca0ef1b1965
2016-02-05 14:34:09 -08:00
James Zern
05437805f7 intrapred/d135: flatten border results before storing
the results along the top and left border are then stored with a moving
window into the vector.
~40-67% faster on ARM, ~40-77+% on x86 depending on the block size.

Change-Id: Iab369aa2946a3ae4eb7290d512868fe5db92dbc8
2016-02-05 12:31:48 -08:00
Alex Converse
def519efb3 Fix an uninitialized warning.
Change-Id: I3ac1070df2948f6d8637ade7192400be53a7675c
2016-02-05 12:06:38 -08:00
James Zern
f7b4364cbb vp9_aq_cyclicrefresh: fix unused-function warning
delete apply_cyclic_refresh_bitrate(). unused since:
3472cbb vp9 aq-mode=3: Keep it on even at low bitrates.

Change-Id: I0fac9a31b59504e31000ac3a8f0b68e8d4320113
2016-02-05 11:56:51 -08:00
Hui Su
e5bd08185a Merge "Add a speed feature for intra filter search" into nextgenv2 2016-02-05 18:49:24 +00:00
Yaowu Xu
105da4128d Fix bad merge artifacts
Temporaly disable warning for unused function for vp10, needs clean
out the warnings before re-enable the flag for vp10.

Change-Id: I5636f8cd607423f6ea6963db9c2cbd688e30b495
2016-02-05 09:04:41 -08:00
Marco Paniconi
85dccb6944 Merge "vp9 aq-mode=3: Keep it on even at low bitrates." 2016-02-05 16:38:41 +00:00
Yaowu Xu
48b2713553 Merge branch 'master' into nextgenv2 2016-02-05 05:00:06 -08:00
James Zern
693dcc86cf Merge "configure: enable -Wunused-function by default" 2016-02-05 02:45:56 +00:00
James Zern
ecd32d6faa Merge "Vidyo patch: Optimization for 1-to-2 downsampling and upsampling." 2016-02-05 02:36:03 +00:00
Marco
3472cbbd88 vp9 aq-mode=3: Keep it on even at low bitrates.
Change-Id: I6a27db6b9dbe3d8cb3d47ef30fbd52557171cadd
2016-02-04 17:31:10 -08:00
Hui Su
9604f69005 Merge "Add 8-tap interpolation filter options for intra prediction" into nextgenv2 2016-02-04 23:06:28 +00:00
Scott LaVarnway
989c69303d Vidyo patch: Optimization for 1-to-2 downsampling and upsampling.
Change-Id: I9cc9780f506e025aea57485a9e21f0835faf173c
2016-02-04 14:50:26 -08:00
Jingning Han
d1c2949eb4 Merge "Generalize the dynamic reference motion vector coding mode" into nextgenv2 2016-02-04 22:28:56 +00:00
Yaowu Xu
027cc8c873 merge per-segment lossless feature in misc_fixes
Change-Id: I56d56781d371c99aa5cdd2db1cbc0a17437723e9
2016-02-04 08:31:16 -08:00
Zoe Liu
0f7dd40324 Define a macro to replace hardcoded values
The definition is for the number of frames to check to determine the
recent decay rate, further to determine the next key frame in the
first pass of the encoder.

Change-Id: Ic696d6eb518a86fa296842273cf8767ef0b0e27a
2016-02-04 05:33:22 -08:00
James Zern
771d276b27 configure: enable -Wunused-function by default
when INLINE is defined and mips is not being targeted. otherwise keep
the old --enable-extra-warnings behavior

Change-Id: Iba576edbe5fca03efa56ce99eee11f9cafc573ad
2016-02-03 23:07:20 -08:00
James Zern
f74e04cc62 fdct8x8_test:: quiet unused-function warning
in non-x86 builds

Change-Id: I9f6fa40d27f95a4afe92444f71c23dc9a15db466
2016-02-03 23:04:42 -08:00
James Zern
497b6744ea Merge changes from topic 'dead-code'
* changes:
  yv12config.c: remove dead CONFIG_ALPHA code
  vp9_encoder.c: remove unused macro
  vpx_mem.c: remove unused macro
  vp9_svc_layercontext.c: relocate a macro definition
  vp9_encoder.c: protect SNPRINT* macros w/CONFIG check
  vp9_aq_360.c: remove some unused macros
  vp9_diamond_search_sad_avx.c: rename (un)likely macros
  vp9_resize.c: add missing include
  vp9_aq_complexity.c: remove unused macros
  vp9_detokenize.c: remove unused macros
2016-02-04 06:25:44 +00:00
Debargha Mukherjee
9909b5414e Merge "Fix for supertx: ignore skip_txfm optimization" into nextgenv2 2016-02-04 04:28:29 +00:00
James Zern
b63788a116 Merge changes I1171e9c7,I77292e11,I1678ab48,Id7a2c345
* changes:
  dct16x16_test: quiet unused-function warning
  intrapred: protect functions w/CONFIG check x2
  dct32x32_test: remove unused function
  tests: quiet some warnings w/--disable-webm-io
2016-02-04 03:43:49 +00:00
Marco Paniconi
37c8f57318 Merge "vp9-skinmap. Some adjustments for model=1." 2016-02-03 23:34:07 +00:00
hui su
5b618b7cae Add a speed feature for intra filter search
Seperate the prediction angle search and fitler search.
It can reduce the computation overhead of filter search by as much
as 85%, while keeping more than 50% of the coding gain.

Change-Id: Id152f71e20ebcaca8b429bdd4ca1fbeb646fc6bf
2016-02-03 15:12:06 -08:00
Yaowu Xu
aac86aa093 Merge "Merge branch 'master' into nextgenv2" into nextgenv2 2016-02-03 22:54:25 +00:00
Marco
82148c02c1 vp9-skinmap. Some adjustments for model=1.
-use larger threshold on y (as in vp8).
-add distance threshold for each cluster
-use larger skin distance threshold for first cluster
-add some early exist checks.

Keep default setting to model=0.

Change-Id: I1044b99ade4bb1f215a860a019a4d84cee2f7715
2016-02-03 14:47:15 -08:00
hui su
3b1c766802 Add 8-tap interpolation filter options for intra prediction
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
2016-02-03 14:19:20 -08:00
Jingning Han
6f49e8a186 Merge "Use precise rate cost estimate for skip block mode" 2016-02-03 20:40:10 +00:00
Jingning Han
4fb8b217e6 Generalize the dynamic reference motion vector coding mode
This commit generalizes dynamic reference motion vector coding mode
to support multiple candidate modes in the rate-distortion
optimization scheme and to support the selection in the bit-stream
syntax. The maximum number of modes allowed is currently limited to
4. The syntax elements for the dynamic reference motion vector
modes are using binary codes. The scheme supports single reference
frame.

It improves the compression performance
derf   0.135%
hevcmr 0.098%

Change-Id: Id053d6ce76e8365e52727bd0d12d28ce3de2e0e8
2016-02-03 11:57:42 -08:00
Jingning Han
bcce658d31 Use precise rate cost estimate for skip block mode
It improves the compression performance of VP9 by 0.1% across all
test sets. No speed change is observed.

Change-Id: I59338c5c9e67bae22188f35fc3afbfe2a6bba6b0
2016-02-03 11:09:16 -08:00
Jingning Han
6cfb488500 Merge "Account for zero-forcing operation in selective ref mv mode" into nextgenv2 2016-02-03 19:07:28 +00:00
Hui Su
2231d1063c Merge "Fix some interger overflow errors" 2016-02-03 18:12:26 +00:00