1. 21 Mar, 2016 3 commits
  2. 18 Mar, 2016 6 commits
    • Sarah Parker's avatar
      Remove prune three from speed features · fab5454a
      Sarah Parker authored
      Not getting good results for this feature, will try again
      when transforms are frozen.
      Change-Id: Id12396786cb9369ad34d0bd845f7beba3a037726
    • Yaowu Xu's avatar
      Two minor logic fixes · 42e5c2ad
      Yaowu Xu authored
      Change-Id: I1d5624fb2f34f87a55613036851034ec7c2d0b76
    • Jingning Han's avatar
      Enable dynamic motion vector referencing for newmv mode · 93748c3e
      Jingning Han authored
      This commit enables the dynamic motion vector predictor for NEWMV
      mode. It allows the codec to select the best motion vector predictor
      in a rate-distortion optimization framework for motion vector
      residual coding. The compression performance is improved:
      lowres  0.14%
      midres  0.27%
      hdres   0.24%
      Change-Id: I6a601c74eb6cb0b71a613336d40363359f2edecd
    • Geza Lore's avatar
      Refactor bsse and skip_txfm in MACROBLOCK. · f8cfb72a
      Geza Lore authored
      Simple refactoring to 2 dimensional arrays, in preparation for 128
      wide superblocks.
      Change-Id: I40d447bd9fbd4f755534ea3cc82fc8f4676cea07
    • Geza Lore's avatar
      Refactor mbmi->inter_tx_size to 2D array. · efe7d4e5
      Geza Lore authored
      This is in preparation of increasing the superblock size.
      Change-Id: I9197e397399fbe8aec1178a45ea0337dd90412d7
    • hui su's avatar
      Minor bug fix in ex-intra experiment · 507c55b2
      hui su authored
      No performance change observed from borg tests.
      Change-Id: I20c232c5dde8cfc84452a4c7185389bd8b812ffd
  3. 17 Mar, 2016 2 commits
  4. 16 Mar, 2016 1 commit
  5. 15 Mar, 2016 2 commits
    • Yaowu Xu's avatar
      Properly set the estimate of rate_nocoef · 2becffae
      Yaowu Xu authored
      This commit fixes the computation of rate_nocoef for situation when
      rate_y is uninitialized at INT_MAX for  x->skip is true.
      Change-Id: If3dde4e4ee16667f4408067d3bb3084f916272f1
    • Angie Chiang's avatar
      remove filter_cache · b6fef124
      Angie Chiang authored
      PSNR test
              lowres  hdres
      lowbd   -0.013  0.067
      highbd  -0.044  0.039
      Change-Id: Iefdb1e966bd004b2027456778185b675e8fb9b81
  6. 14 Mar, 2016 1 commit
    • Yue Chen's avatar
      Speed up rd selection in OBMC experiment · b5f8b70c
      Yue Chen authored
      Instead of testing all interpfilter-BMC/OBMC combinations, we choose
      the best interpolation filter based on regular inter prediction.
      Reduction in encoding time: ~10%
      Drop in performance gain: 0.08% lowres, 0.04% midres
      Change-Id: Ifc19097a918ac76b529db9af4c60e2c70e93f7ad
  7. 12 Mar, 2016 1 commit
  8. 11 Mar, 2016 2 commits
  9. 10 Mar, 2016 2 commits
  10. 09 Mar, 2016 3 commits
  11. 08 Mar, 2016 1 commit
  12. 07 Mar, 2016 1 commit
  13. 04 Mar, 2016 1 commit
    • Sarah Parker's avatar
      Adding speed feature interface for ext tx search · 2ca7d42e
      Sarah Parker authored
      This sets up the interface for 3 speed features that progressively
      eliminate a greater number of transforms in ext tx using
      pre-trained support vector machines.
      Each speed feature still needs to be implemented.
      Change-Id: Ia508aeadc0cffdc080fb227f357a5d1dfbca08e2
  14. 03 Mar, 2016 1 commit
  15. 01 Mar, 2016 1 commit
    • Yaowu Xu's avatar
      Fix an overflow issue for HBD · 5c613ea8
      Yaowu Xu authored
      The sum of squared value of a block can overflow 32bit, this commit
      changes to use int64_t to avoid the overflow issue.
      Change-Id: I78fcd6999634f186f86d649cfce85d97a993d040
  16. 29 Feb, 2016 1 commit
    • Yunqing Wang's avatar
      Do sub-pixel motion search in up-sampled reference frames · 342a368f
      Yunqing Wang authored
      Up-sampled the reference frames to 8 times in each dimension using
      the 8-tap interpolation filter. In sub-pixel motion search, use the
      up-sampled reference frames to find the best matching blocks. This
      largely improved the motion search precision, and thus, improved
      the compression quality. There was no change in decoder side.
      Borg test and speed test results:
      1. On derflr set,
      Overall PSNR gain: 1.306%, and SSIM gain: 1.512%.
      Average speed loss on derf set was 6.0%.
      2. On stdhd set,
      Overall PSNR gain: 0.754%, and SSIM gain: 0.814%.
      On hevchd set,
      Overall PSNR gain: 0.465%, and SSIM gain: 0.527%.
      Speed loss on HD clips was 3.5%.
      Change-Id: I300ebaafff57e88914f3dedc8784cb21d316b04f
  17. 26 Feb, 2016 2 commits
    • Debargha Mukherjee's avatar
      Some refactoring and cleanups of interp filter · bab2912b
      Debargha Mukherjee authored
      Includes various cosmetic changes and refactoring including
      naming the sharp filters differently (since they are no longer
      Change-Id: Ida5a19ca0daa9f6a64a6734394c685b2a4a2564a
    • Geza Lore's avatar
      Port interintra experiment from nextgen. · 7ded038a
      Geza Lore authored
      The interintra experiment, which combines an inter prediction and an
      inter prediction have been ported from the nextgen branch. The
      experiment is merged into ext_inter, so there is no separate configure
      option to enable it.
      Change-Id: I0cc20cefd29e9b77ab7bbbb709abc11512320325
  18. 24 Feb, 2016 2 commits
  19. 23 Feb, 2016 2 commits
    • Yue Chen's avatar
      Optimizing obmc rd decision by checking the real rd cost · a614262e
      Yue Chen authored
      Instead of using model_rd_for_sb() to estimate the cost and make the
      decision on bmc/obmc, we use super_block_yrd/uvrd() to calculate and
      compare the real rd costs of bmc and obmc.
      Average bit-rate reduction(%) of obmc experiment:
      Before the optimization, the coding gain was:
      Note: there is still some mysterious bug because that compared to
      the previous version, the performance at low bit rate drops a lot.
      Change-Id: I8dbee04a272190f10516a3953c1ae690f8136766
    • Geza Lore's avatar
      Experiment to use image domain dist in baseline. · 3c4b56c4
      Geza Lore authored
      Change-Id: Ib29f510289716b5ab5c7d74d32a450c190308a83
  20. 20 Feb, 2016 2 commits
  21. 19 Feb, 2016 1 commit
  22. 18 Feb, 2016 2 commits
    • Yaowu Xu's avatar
      Properly normalize HBD sse computation · 0c0f3efd
      Yaowu Xu authored
      This fixes a bug in HBD sum of squared error computation introduced
      in  #abd00505.
      Change-Id: I9d4e8627eb8ea491bac44794c40c7f1e6ba135dc
    • Julia Robson's avatar
      cost_coeff speed improvements · c6eba0b4
      Julia Robson authored
      Preliminary tests indicated that these changes make cost_coeffs
      approximately 20% faster which is a 2% improvement overall
      Change-Id: Iaf013ba75884415cd824e98349f654ffb1c3ef33