summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2015-03-16Merge "Remove duplicate clamping"Yaowu Xu
2015-03-16Merge "Remove ineffective newmv skip checking from vp9_pick_inter_mode"Jingning Han
2015-03-16Merge "Simplify prediction filter search in rtc coding mode"Jingning Han
2015-03-16Merge "Refactor column integral projection computation"Jingning Han
2015-03-16Merge "change the order of inter modes evaluated"Yaowu Xu
2015-03-16Remove duplicate clampingYaowu Xu
The mvs are clamped in the vp9_find_best_ref_mvs() already. Change-Id: I9bea5e35aef6007466fe7fca4bc2dc5c17e74222
2015-03-16Remove ineffective newmv skip checking from vp9_pick_inter_modeJingning Han
Change-Id: I41ee684cf113a7b5edf280183e51cb08b2e93cc4
2015-03-16Simplify prediction filter search in rtc coding modeJingning Han
Reduce unnecessary fetch from MB_MODE_INFO. Change-Id: Iff89b76d5e2774c00a564e902913a633fa2e1ea9
2015-03-16change the order of inter modes evaluatedYaowu Xu
Change-Id: I10c1ad23b110cf92cb026e895039c215c47abfd0
2015-03-16Merge changes I67890f52,I33c72799James Zern
* changes: usage.dox: fix encoder/decoder subpage list indent usage.dox: fix doxygen warnings in 1.8.x
2015-03-16Refactor column integral projection computationJingning Han
Move the scaling factor outside column projection. This avoids repeated calculation of the same scaling factor. Profiling shows that the percentage of vp9_int_pro_col_sse2 of overall cycles goes from 2.29% down to 1.88%. Change-Id: I5ac4e324ab2d7f33ba2de66dd2a12e04e04dfd66
2015-03-16Merge "Fix indent in choose_partitioning"Jingning Han
2015-03-16Fix indent in choose_partitioningJingning Han
Change-Id: I4039f8ac75a9cfcc4d07abd0619d1379bb10fe51
2015-03-16vp9_pick_inter_mode(): minor optimizationsYaowu Xu
1. remove duplicate initialization to mbmi->interp_filter. 2. move mv clamping into ref_frame loop instead of mode checking loop. 3. move the check if last frame is same as golden frame earlier to avoid initialization of Golden reference related variables. Change-Id: Idf2d05e19e94a24f69cc289687869fc71d2ff289
2015-03-15Merge "Fix choose_partitioning threshold setup for speed -5"Jingning Han
2015-03-14Fix 1-step refinement search tableJingning Han
Change-Id: I32f0bcb40c6e7ba63bfae487739ededd0b6b2dde
2015-03-13usage.dox: fix encoder/decoder subpage list indentJames Zern
use \li to separate the list items contained in conditionals. this avoids the encode page becoming a sub-item of decode; likely a problem in <1.8.3. + fix encoder conditional, spelling error + correct encode page name to match decode 'Encoding' Change-Id: I67890f52bed8e708bad63fb8819a074e0beff2ca
2015-03-13usage.dox: fix doxygen warnings in 1.8.xJames Zern
use \li to denote list items with \if. fixes the following likely visible in <1.8.3: usage.dox: warning: Invalid list item found usage.dox: warning: End of list marker found without any preceding list items Change-Id: I33c72799edf9f8866596ac8f79247050b8c75681
2015-03-13Fix choose_partitioning threshold setup for speed -5Jingning Han
The compression performance of speed -5 is on average 12.6% better than speed -6. At lower bit-rates, the gains are typically 20% or more. For 2-thread encoding, the speed -5 takes about 1.6x time of speed -6. Change-Id: If7a73464a24d33e8f49b9533b51ec51c8da7fc80
2015-03-13Merge "Fix comments for vpx_codec_enc_config_default()"Yaowu Xu
2015-03-13Fix comments for vpx_codec_enc_config_default()Yaowu Xu
This makes clear that for end-usage, only 0 is supported currently. Change-Id: I99795807c20df1d71b3c803e8d5f3b36583dcc85
2015-03-13Merge "Provide information on codec controls"Yaowu Xu
2015-03-13Merge "Set default encode test driver to 1 column and 1 thread."Frank Galligan
2015-03-13Merge "Fix crash with vp9 denoiser on."Marco
2015-03-13Merge "Use sdx4df to do 1-step refinement"Jingning Han
2015-03-13Provide information on codec controlsYaowu Xu
The commit updates the comments in vp8cx.h to make it clear which codec support each of codec control functions. Change-Id: Ibf876e289d4325bbb61ce19311da60d384624c2f
2015-03-13Fix crash with vp9 denoiser on.Marco
Crash occured on very first key frame, because denoiser temporal function was beng entered. Updated denoiser unittest to set cpu_used from first frame, and verified fix fixes the crash. Change-Id: I3be1124b52846fbbe7248d2c3d6136e086c80bc1
2015-03-13Merge "Lower bitrate threshold below which cyclic refresh is turned off."Marco
2015-03-13Merge "Reconcile active_map and cyclic refresh"Alex Converse
2015-03-13Merge "Reset src buffer only once in vp9_int_pro_motion_estimation"Jingning Han
2015-03-13Merge "Reduce the number of full block SAD calls"Jingning Han
2015-03-13Use sdx4df to do 1-step refinementJingning Han
Change-Id: Ie0c3ef3ae3aedf049b1a296de607730b79c12672
2015-03-13Lower bitrate threshold below which cyclic refresh is turned off.Marco
Change-Id: Ib54ab11adf8178eec74f65388a89c8f912c7869a
2015-03-13Merge "Shorten GF/arf interval in hard scenes."paulwilkins
2015-03-13Set default encode test driver to 1 column and 1 thread.Frank Galligan
Change-Id: I70ee538609cefad76b8ce9b225bbf8d8447aff23
2015-03-12Merge changes Ic54c88ec,I6f120b65Yaowu Xu
* changes: Added comment for vp9 only codec controls Rename a vp9 only control with vp9e prefix
2015-03-12Reset src buffer only once in vp9_int_pro_motion_estimationJingning Han
Change-Id: I5c96b6a25f9df60da65b7af7c92a921b611746e3
2015-03-12Merge "vp9_pick_inter_mode(): Use single loop to evaluate inter modes"Yaowu Xu
2015-03-12Added comment for vp9 only codec controlsYaowu Xu
Comments are updated to reflect that these controls apply to VP9 only, thereby, to insure the document produced by doxygent to reflect the same fact too. Change-Id: Ic54c88ec066aa0ec4552d43dd4a7016e1f810f42
2015-03-12Rename a vp9 only control with vp9e prefixYaowu Xu
Also updated comments and doc to make it clear that the control is supported in VP9 only. Change-Id: I6f120b65135fc1185695e9e3dc29fbe0d430824b
2015-03-12Reduce the number of full block SAD callsJingning Han
This commit uses a 6-point 1-step refine motion search in the integral projection based full pixel motion estimation, to replace the current 9-point search. It reduces runtime cost of speed -6 on some noisy clips, e.g., dark720p single thread 33314 b/f, 40.076 dB, 18231 ms -> 33307 b/f, 40.067 dB, 17768 ms The compression performance for rtc set remains unchanged. Change-Id: I194ea5a9ce52e5a10baeee36338633adc22f764c
2015-03-12Merge "Minorly modify model_rd_for_sb_y function"Yunqing Wang
2015-03-12Merge "Fix fdct8x8_quant ssse3 overflow issue"Jingning Han
2015-03-12Reconcile active_map and cyclic refreshAlex Converse
Change-Id: Id7f8654aeeb20caa402bc822521b1d72c658f4f9
2015-03-12vp9_pick_inter_mode(): Use single loop to evaluate inter modesYaowu Xu
This commit changes to use single loop to evaluate all inter modes. There is no impact on compression quality and speed, but allow future experiment with the order of modes evaluated. Change-Id: I71696ce1014cbe127e25e98710d835987f5ecc09
2015-03-12Minorly modify model_rd_for_sb_y functionYunqing Wang
Added a skip_dc check. If skip_dc = 1, we could eliminate calling of vp9_model_rd_from_var_lapndz(). This gave slight PSNR & SSIM gain(<0.1%), and no speed change. Change-Id: If5ca733366148c86b98e196a00cc890f50e9a3e5
2015-03-12Fix fdct8x8_quant ssse3 overflow issueJingning Han
This resolves webm issue 968. Change-Id: Ieb363129b1e135a561141c68211d413226aba754
2015-03-12Merge "Some rate control adjustments to control overshoot"Deb Mukherjee
2015-03-12Merge "Prevent integer overflow in choose_partitioning"Jingning Han
2015-03-12Merge "Remove unnecessary speed feature checking"Jingning Han
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-11 03:19:05 +0000