shithub: libvpx

3861b25b – Marco <marpan@google.com> authored on 2016/04/07 13:20

vp9: Fix to active_best for GF/ARF in 1 pass vbr.

5b3d3b19 – Alex Converse <aconverse@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/11 14:58

Merge "Remove obsolete segment skip checks from tokenization."

f659c7e9 – Paul Wilkins <paulwilkins@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/11 06:27

Merge "Adjust motion component of prediction decay."

c5a89b46 – Paul Wilkins <paulwilkins@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/11 06:27

Merge "Trap very short arf group just before a kf."

ad47d1d1 – Scott LaVarnway <slavarnway@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/10 18:17

Merge "VP9: Combine TileData with TileWorkerData"

4317ba58 – Tom Finegan <tomfinegan@google.com> authored on 2016/03/24 09:12

third_party: Roll libwebm snapshot.

985786c7 – Tom Finegan <tomfinegan@google.com> authored on 2016/04/08 13:10

configure: Disable webm_io when target is VS 7/8/9.

6a0e7715 – Hui Su <huisu@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/08 14:03

Merge "Correct comments for scan order neighbors"

989d5368 – Marco Paniconi <marpan@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/08 13:51

Merge "vp9-denoiser: Avoid copy-block when denoising is at LowLow level."

86212bc3 – Marco Paniconi <marpan@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/08 13:51

Merge "vp8-denoiser: Avoid doing the mcomp if we don't denoise."

fa4ca403 – Alex Converse <aconverse@google.com> authored on 2016/04/08 07:38

Remove obsolete segment skip checks from tokenization.

81d1e7bf – paulwilkins <paulwilkins@google.com> authored on 2016/04/08 06:28

Adjust motion component of prediction decay.

13ce6e6d – Marco Paniconi <marpan@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/07 18:00

Merge "vp9: Some adjustments to scene-cut detection."

107de196 – paulwilkins <paulwilkins@google.com> authored on 2016/04/07 10:55

Trap very short arf group just before a kf.

f6f54427 – Marco <marpan@google.com> authored on 2016/04/07 07:57

vp9: Some adjustments to scene-cut detection.

e96f0c3d – Marco Paniconi <marpan@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/07 12:31

Merge "vp9-1 pass vbr: Use average QP for active_worst_quality, for inter-frames."

6965bde4 – Tom Finegan <tomfinegan@google.com> authored on 2016/04/07 07:47

Revert "third_party: Roll libwebm snapshot."

7cb4ab56 – Scott LaVarnway <slavarnway@google.com> authored on 2016/04/07 07:44

VP9: Combine TileData with TileWorkerData

69c7ad34 – hui su <huisu@google.com> authored on 2016/04/07 07:04

Correct comments for scan order neighbors

31b97571 – Tom Finegan <tomfinegan@google.com> authored on 2016/03/24 09:12

third_party: Roll libwebm snapshot.

758fc662 – Tom Finegan <tomfinegan@google.com> authored on 2016/04/06 10:26

Revert "third_party: Roll libwebm snapshot."

c93946bb – Scott LaVarnway <slavarnway@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/06 16:51

Merge "VP9: refactor read_switchable_interp_filter()"

99211b02 – Tom Finegan <tomfinegan@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/06 11:03

Merge "Fix crash when mkvparser returns a NULL BlockEntry."

25920137 – Scott LaVarnway <slavarnway@google.com> authored on 2016/04/06 00:53

VP9: refactor read_switchable_interp_filter()

5ab46e0e – James Zern <jzern@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/05 22:51

Merge changes I7a1c0cba,Ie02b5caf,I2cbd85d7,I644f35b0

16d6aace – Yi Luo <luoyi@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/05 14:26

Merge "Fix high bit depth mask and variance reference function"

4bdc7d1c – Yi Luo <luoyi@google.com> authored on 2016/04/04 05:44

Fix high bit depth mask and variance reference function

74aaa238 – Tom Finegan <tomfinegan@google.com> authored on 2016/03/24 09:12

third_party: Roll libwebm snapshot.

ac86dde3 – Marco <marpan@google.com> authored on 2016/04/04 07:05

vp9-denoiser: Avoid copy-block when denoising is at LowLow level.

38bc1d0f – James Zern <jzern@google.com> authored on 2016/04/01 08:40

vpx_fdct16x16_1_sse2: improve load pattern

eb64ea3e – James Zern <jzern@google.com> authored on 2016/03/29 17:04

vpx_fdct16x16_1_c/msa: fix accumulator overflow

3735def6 – James Zern <jzern@google.com> authored on 2016/04/01 08:35

vpx_fdctNxN_1_sse2: reduce store size

6877acb6 – Marco <marpan@google.com> authored on 2016/04/04 09:00

vp8-denoiser: Avoid doing the mcomp if we don't denoise.

71cf9fac – Marco Paniconi <marpan@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/04 13:33

Merge "vp9-noise estimation: Increase threshold for Low-level."

5572ab1d – Marco <marpan@google.com> authored on 2016/04/04 08:30

vp9-noise estimate: Adhust threshold for Low level detection.

79d5aac7 – Tom Finegan <tomfinegan@google.com> authored on 2016/04/04 07:49

Fix crash when mkvparser returns a NULL BlockEntry.

9ff73fe0 – Marco <marpan@google.com> authored on 2016/04/04 05:25

vp9-noise estimation: Increase threshold for Low-level.

a2307ff0 – Marco <marpan@google.com> authored on 2016/04/04 05:11

vp9-denoiser. Code cleanup and bugfix.

89b1c9d4 – James Zern <jzern@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/02 15:38

Merge "dct32x32_test: create fn pointers w/'&' ref"

50e05d06 – James Zern <jzern@google.com> authored on 2016/04/02 07:17

dct32x32_test: create fn pointers w/'&' ref

c21d4370 – James Zern <jzern@google.com> authored on 2016/04/02 07:04

vpx_fdct32x32_1_msa: fix accumulator overflow

a0359b8c – James Zern <jzern@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/01 19:12

Merge changes I8e9692fc,Ifc06110c

2f0067c6 – Marco Paniconi <marpan@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/04/01 17:26

Merge "vp9-denoiser: Adjust usage of skin and consec_zeromv."

c98f8e04 – James Zern <jzern@google.com> authored on 2016/04/01 15:44

dct32x32_test: add PartialTrans32x32Test, Random

f4cae05c – James Zern <jzern@google.com> authored on 2016/04/01 08:21

vpx_fdctNxN_1_c: remove unnecessary store

7467ee04 – Marco <marpan@google.com> authored on 2016/04/01 07:41

vp9-denoiser: Adjust usage of skin and consec_zeromv.

bc6c1997 – Marco <marpan@google.com> authored on 2016/01/19 07:57

vp8 denoiser: Some adjustments to usage of skin and motion.

a2a97b86 – Scott LaVarnway <slavarnway@google.com> authored on 2016/03/29 11:03

VP9: Refactor vp9_decode_block_tokens()

74ed95a3 – James Zern <jzern@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/31 23:33

Merge "disable vp9_diamond_search_sad_avx"

569c0dfc – James Zern <jzern@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/31 23:31

Merge "vpx_ports/mem_ops.h: clear some -Wconversion warnings"

4aac819a – Debargha Mukherjee <debargha@google.com> authored on 2016/03/30 09:53

Fix some static analysis issues with resize

18583611 – Scott LaVarnway <slavarnway@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/30 19:21

Merge "VP9: Eliminate up_available and left_available"

8fab3e82 – Marco <marpan@google.com> authored on 2016/03/30 12:27

vp9-1 pass vbr: Use average QP for active_worst_quality, for inter-frames.

057c1c40 – James Zern <jzern@google.com> authored on 2016/03/30 11:28

disable vp9_diamond_search_sad_avx

44a9edf4 – James Zern <jzern@google.com> authored on 2016/03/30 13:02

vpx_ports/mem_ops.h: clear some -Wconversion warnings

0269df41 – James Zern <jzern@google.com> authored on 2016/03/29 17:04

vpx_fdct32x32_1_c: fix accumulator overflow

8d8ee1f6 – James Zern <jzern@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/30 01:38

Merge "vp8_peek_si_internal: quiet static analysis warning"

ba962a5f – Scott LaVarnway <slavarnway@google.com> authored on 2016/03/30 00:47

VP9: Eliminate up_available and left_available

904cb533 – Johann Koenig <johannkoenig@chromium.org> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/29 18:36

Merge "Properly propagate out of memory errors."

aea8d97a – Paul Wilkins <paulwilkins@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/29 10:02

Merge "Increase min-max q range for normal inter frames."

6f51672c – Jean-Yves Avenard <jyavenard@mozilla.com> authored and Johann <johannkoenig@google.com> committed on 2015/07/11 16:47

Properly propagate out of memory errors.

48e7294e – Marco <marpan@google.com> authored on 2016/03/24 11:29

vp9-denoiser: Incorporate consec_zeromv in skin detection for denoising.

86fb8130 – Marco Paniconi <marpan@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/28 12:05

Merge "vp8-denoiser: Use the same skin detection for model=1 as in vp9."

ee1bd86d – James Zern <jzern@google.com> authored on 2016/03/26 08:52

Revert "third_party: Roll libwebm snapshot."

aa1132d1 – Tom Finegan <tomfinegan@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/25 21:41

Merge "third_party: Roll libwebm snapshot."

f94124cf – Marco <marpan@google.com> authored on 2016/03/25 07:34

vp9: 1 pass vbr mode at speed 5: switch to use mv.search to NSTEP.

80ddeb28 – Tom Finegan <tomfinegan@google.com> authored on 2016/03/24 09:12

third_party: Roll libwebm snapshot.

1bf7a0fb – James Zern <jzern@google.com> authored on 2016/03/25 05:57

vp8_peek_si_internal: quiet static analysis warning

79dc6c03 – James Zern <jzern@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/24 18:04

Merge changes Id0a490bc,I1b4a0504

04b4a5f2 – Paul Wilkins <paulwilkins@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/24 14:05

Merge "Force recode for bad rate misses."

27a6be14 – Paul Wilkins <paulwilkins@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/24 14:04

Merge "Fix some HBD scaling bugs."

5f90713b – Marco <marpan@google.com> authored on 2016/03/24 11:42

vp8-denoiser: Use the same skin detection for model=1 as in vp9.

82762e24 – Scott LaVarnway <slavarnway@google.com> authored and James Zern <jzern@google.com> committed on 2016/02/10 00:06

VP9: Eliminate extra mv clamp in decoder

715d3bbd – James Zern <jzern@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/23 23:57

Merge "Add a TODO for using tile rows while encoding with multiple threads"

c20c955e – James Zern <jzern@google.com> authored on 2016/03/23 16:32

vp9/inverse_transform_block_inter: move eob check

f64a30ac – James Zern <jzern@google.com> authored on 2016/03/23 16:24

vp9/inverse_transform_block_intra: move eob check

813221a1 – Johann Koenig <johannkoenig@chromium.org> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/23 16:04

Merge "Allows sub-second frame rates in vp9"

a21fc45b – Yunqing Wang <yunqingwang@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/23 11:52

Merge "Prevent encoder crash caused by row tile dependencies"

5486e1d8 – Yunqing Wang <yunqingwang@google.com> authored on 2016/03/23 08:23

Add a TODO for using tile rows while encoding with multiple threads

c8fc6ed1 – Marco Paniconi <marpan@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/22 21:06

Merge "vp9: Non-rd partition selection: use consec_zeromv to set sb_is_skin."

a6246927 – Yaowu Xu <yaowu@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/22 19:45

Merge "Simplify the loopfilter synchronization logic in VP8 encoder"

dc5f7712 – Yunqing Wang <yunqingwang@google.com> authored on 2016/03/22 06:54

Prevent encoder crash caused by row tile dependencies

b198bcd5 – Yunqing Wang <yunqingwang@google.com> authored on 2016/03/22 10:13

Simplify the loopfilter synchronization logic in VP8 encoder

a84f1f84 – Paul Wilkins <paulwilkins@google.com> authored on 2016/03/22 08:56

Increase min-max q range for normal inter frames.

9d8ead12 – Paul Wilkins <paulwilkins@google.com> authored on 2016/03/15 08:54

Force recode for bad rate misses.

ee35f4de – Paul Wilkins <paulwilkins@google.com> authored on 2016/03/18 06:25

Fix some HBD scaling bugs.

01283e27 – Marco <marpan@google.com> authored on 2016/03/21 13:10

vp9: Non-rd partition selection: use consec_zeromv to set sb_is_skin.

bfc2a7e3 – Marco Paniconi <marpan@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/18 13:58

Merge "vp9: Improvement to skin detection."

a3562b60 – Marco <marpan@google.com> authored on 2016/03/18 04:38

vp8-denoiser: Update on the sum_diff threshold.

e1ad39d7 – James Zern <jzern@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/17 22:38

Merge "altref_test: mark AltRefForcedKeyTest as large"

0dcd21fb – James Zern <jzern@google.com> authored on 2016/03/17 14:46

altref_test: mark AltRefForcedKeyTest as large

ce39add4 – Marco <marpan@google.com> authored on 2016/03/14 07:14

vp9: Improvement to skin detection.

32c38a20 – Marco <marpan@google.com> authored on 2016/03/17 08:17

vp9: Fix the threshold for gf update in cyclic refresh.

e6aebcdf – Alex Converse <aconverse@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/17 12:25

Merge "Use whole pixel only at speed 8 screen content."

54e5ff00 – Alex Converse <aconverse@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/17 12:25

Merge "Reuse computation in build_nmv_component_cost_table."

ed5d2e71 – Alex Converse <aconverse@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/17 12:24

Merge "Fix using screen content tuning in all the cpu speed tests."

825e7542 – Marco Paniconi <marpan@google.com> authored and Gerrit Code Review <noreply-gerritcodereview@google.com> committed on 2016/03/17 11:14

Merge "vp9-svc: Bugfix on usage of svc_force_zeromv in non-rd pickmode."

e17f6347 – Alex Converse <aconverse@google.com> authored on 2016/03/16 17:57

Fix using screen content tuning in all the cpu speed tests.

0271bafb – Alex Converse <aconverse@google.com> authored on 2016/03/16 11:10

Reuse computation in build_nmv_component_cost_table.

bd924fbd – Alex Converse <aconverse@google.com> authored on 2016/03/16 11:56

choose_partitioning: Avoid computing segment_id twice.