summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2016-04-26Merge "Add the 64-bit CPU cycle count utility function"Yi Luo
2016-04-26Merge "Resolve a couple of TODOs in firstpass.c"Paul Wilkins
2016-04-26Merge "Experiment to adapt for net AQ offset."Paul Wilkins
2016-04-25vp9: Fix consec_zero_mv in skin detection.Marco
This was missed in https://chromium-review.googlesource.com/#/c/340135/. Change-Id: Iafc83141f4ff744819bdbde7d239678b215feb9d
2016-04-25Merge "vp9: 1 pass vbr: Adjust gf boost/interval on motion level."Marco Paniconi
2016-04-25Merge "vp9: Fix a bug where best_sse_mv is not set properly."Jacky Chen
2016-04-25Merge "vpx_minmax_8x8_neon and test"Johann Koenig
2016-04-25vp9: 1 pass vbr: Adjust gf boost/interval on motion level.Marco
Add some adjustment for high motion case. Change-Id: If700be1ce849c187ed031221ea7cabbc8d91ec2b
2016-04-25vp9: Fix a bug where best_sse_mv is not set properly.JackyChen
In VP9 internal denoiser, motion magnitude is computed from best_sse_mv, which should be set to 0 at the begining. This bug may cause visual aritifact in denoiser. Also, delete two improper comments. Change-Id: I8710d2acba23320bc85cf72af17d65245c19438b
2016-04-25Merge "vp9-denoiser: Bugfix and some adjustments for high noise case."Marco Paniconi
2016-04-25vp9-denoiser: Bugfix and some adjustments for high noise case.Marco
Need to check that sse for non-zero mv has been set for the current block (i.e., check that nonzero-mv is tested as a mode, so newmv_sse != UINT_MAX) before forcing to not use zero-mv for denoising. Also increase some thresholds (sse and sse_diff) for high noise case, and use shift operaton instead of multiplication on a threshold computation. Change-Id: Iae7339475d57240316b7fa8b887c4ee3c0d0dbec
2016-04-25vp9: Add rc quantity to track amount of low motion in scene.Marco
Use it for now in noise estimation to bypass estimation if motion level is high. Change-Id: I033662dc909f2060e4e81abf562a7ad262dc8170
2016-04-25Merge "VP9: Do not call vp9_adjust_mask() in vp9_setup_mask()"Scott LaVarnway
2016-04-25Resolve a couple of TODOs in firstpass.cpaulwilkins
Resolved two TODO items. Force a minimum value of 1.0 for frame duration as per section duration. Column inactive zone is currently set to 0 as most of the serious issues relating to inactive regions relate to letter boxing. Change-Id: Ifbab3acf2c089d7305620a7ff7ed7c3536cc9235
2016-04-25Experiment to adapt for net AQ offset.paulwilkins
In Aq mode 1 the segment and AQ delta for each block is based on spatial variance. There may be a net imbalance between blocks that have lower Q than the baseline value and those that have higher Q. This patch monitors that imbalance and extends the allowed baseline Q range for the frame to accommodate adjustment of that baseline value to compensate. Change-Id: Iae8a48c7c01fe2af94a141e149d03acf467237ca
2016-04-22VP9: Do not call vp9_adjust_mask() in vp9_setup_mask()Scott LaVarnway
vp9_adjust_mask() is called again in loop_filter_rows(). Change-Id: If52f5339dfa7971c47b12f9e05f87951044d9391
2016-04-22Merge "hadamard_test: align src/dst buffers"James Zern
2016-04-22Merge "vp9: Some adjustments to denoiser under high-noise condition."Marco Paniconi
2016-04-22vp9-noise estimation: Add check on whether skin detection is enabled.Marco
Change-Id: Ie80b7a14e3505c40b5742793f84aa6fb9d21dab5
2016-04-22vp9: Some adjustments to denoiser under high-noise condition.Marco
Change-Id: Ib47d742d07f13c7c39cea85e228c19ae298baf4a
2016-04-22Merge "vp9: Move consec_zero_mv from cyclic refresh to cpi struct."Marco Paniconi
2016-04-22Add the 64-bit CPU cycle count utility functionYi Luo
Change-Id: Ie87245bbdf5735bc9729199eeb07899d81dbf267
2016-04-22vp9: Move consec_zero_mv from cyclic refresh to cpi struct.Marco
So it can be used even with aq-mode=3 not enabled. Also cleans up some code in the places where its used. No change in behavior. Change-Id: Ib6b265308dbd483f691200da9a0be4da4b380dbc
2016-04-22hadamard_test: align src/dst buffersJames Zern
fixes crash in 32-bit builds Change-Id: I149e517c940a8cd43643e213bfbe077b6f579090
2016-04-21vpx_minmax_8x8_neon and testJohann
BUG=https://bugs.chromium.org/p/webm/issues/detail?id=1156 Change-Id: Ief0ad8d6255b0ef0f233cda153799e3c72d3dbc6
2016-04-22Merge "hadamard 8x8 test"Johann Koenig
2016-04-21hadamard 8x8 testJohann
The order of the output structure is not currently important. BUG=https://bugs.chromium.org/p/webm/issues/detail?id=1021 Change-Id: Ibc0006d569675db6c5060c4529f5d9e73f2e96a6
2016-04-21Merge "vp9_loopfilter.c : fix / clarify todo"James Bankoski
2016-04-21Merge "vp9_loopfilter.c: Todo clean up encoder should work like decoder."James Bankoski
2016-04-21Merge "vp9_loopfilter.c : todo cleanup"James Bankoski
2016-04-21vp9_loopfilter.c : fix / clarify todoJim Bankoski
Change-Id: Ie3ec67a83d1877d3deae9c7922b6899d915aa19e
2016-04-21vp9_loopfilter.c: Todo clean up encoder should work like decoder.Jim Bankoski
Change-Id: I570c6859d6e18cd94ce4a29068477b937489399c
2016-04-21vp9_encodemb.c: TODO clean upJim Bankoski
huisu did in nextgen branch -> please try in vp9 Change-Id: I0ff35db07ac38464e0e2858e303be686c03a5d0e
2016-04-21vp9_loopfilter.c : todo cleanupJim Bankoski
Removed this todo because of another todo which says none of this code should exist. It should be integrated into the block by block encode process as per the decoder. Change-Id: I076bd15140a060e69c014dd7d7cd07fea260aba3
2016-04-21Merge "vp9_loopfilter.c: Todo try inloop calculation."Yaowu Xu
2016-04-20vp9_loopfilter.c: Todo try inloop calculation.Jim Bankoski
This is implemented in the decoder already. Will add a todo for the encoder. Change-Id: I5e78c045cb2edb5ba171022aeeb70051a708b916
2016-04-20vp9: 1 pass vbr: Small fix on reset avg_intersize_gfint on scene cut.Marco
Change-Id: Ie73d1bbc2c953d010b32097440698f0d060750be
2016-04-20vp9: 1 pass vbr: small adjustment to active_best factor for inter.Marco
Change-Id: I9c68aa9e72969b0ffbf57eb186380ff666745aeb
2016-04-20Merge "vp8: Update on sum_diff_threshold."Marco Paniconi
2016-04-19vp9: Adjust gf period for high average Q and overshoot.Marco
For 1 pass vbr mode. Increase the gf interval for case where average Q is close to max and high overshoot is detected. Small increase in overall avg_psnr/sssim metrics (~0.2/0.1%) for ytlive, but improves the low-end (low bitrate) for several clips (less overshoot). Change-Id: Ifba40f25b4861b2e0d9832c82d5359a6a3dce9f2
2016-04-19vp8: Update on sum_diff_threshold.Marco
Reset to previous values. Change-Id: Ia49bafc1d1ec5f657819ffba04d11fd701f315e8
2016-04-18vp9: 1 pass vbr: More even spacing for gf near key frame.Marco
More even spacing near key frame and avoid gf on scene cut if its close to key frame. Small increase in metrics for ytlive set (which uses key-period=150). (~0.2% gain) Change only affects 1 pass vbr mode. Change-Id: If1e5a59baf1e0befbaf998522fbc47d94ac5b5df
2016-04-18vp9: Adjustment to active_best_quality for inter_frame, 1 pass vbr.Marco
Change only affects 1 pass vbr. Use a q value somewhat larger (~6%) than avg_frame_qindex[INTER] as basis for active_best_quality for inter-frames. And use the minium of this (avg_frame_qindex) and the active_worst_quality. This reduces some overshoot in ytlive clips. Overall small but positive average increase in metrics (up on average ~0.2%). Change-Id: Icdbaae7872d5675fd38a13c0ec6ce0e2e3b919ce
2016-04-18Merge "Adjustment to prediction decay."Paul Wilkins
2016-04-15Merge "VP9: inline vp9_get_intra_inter_context()"Scott LaVarnway
2016-04-15Merge "Enable vpx_idct32x32_1024_add_neon for neon as well, not only for ↵Johann Koenig
neon_asm"
2016-04-15VP9: inline vp9_get_intra_inter_context()Scott LaVarnway
Change-Id: I71366140799b9b39474b9b459082cdb250bd1905
2016-04-15Enable vpx_idct32x32_1024_add_neon for neon as well, not only for neon_asmMartin Storsjo
This was never hooked up for the 32x32_34 case as the neon_asm version in 3f7c12da, when the intrinsics version was added. Change-Id: Ic7db4ce5850c637315f9fe9e2de93a4f8cf9e320
2016-04-14Apply 'const' to data not pointerJohann
Change-Id: Ic6b695442e319f7582a7ee8e52a47ae3e38c7298
2016-04-14Merge "dct32x32_test: s/HAVE_NEON_ASM/HAVE_NEON/"James Zern