1. 07 Sep, 2017 1 commit
    • Debargha Mukherjee's avatar
      Reduce line buffer size for Wiener filter. · 22bbe4cc
      Debargha Mukherjee authored
      This patch forces the vertical filtering for the top and bottom
      rows of a processing unit for the Wiener filter to be 5-tap.
      The 5-taps are derived from the primary 7-tap fitler by forcing
      the taps at the end to be zero, and absorbing their weights into
      the other taps to maintain normalization.
      This will effectively reduce the line buffer size for luma Wiener
      filter to 4 (from 6).
      
      Change-Id: I5e21b58369777eabf553a8987387d112f98a5598
      22bbe4cc
  2. 06 Sep, 2017 2 commits
    • Yaowu Xu's avatar
      Silent compiler warning of unused variable · 069cc313
      Yaowu Xu authored
      Change-Id: I1b921e79484e78a272c4f62e30cb85631a65ac25
      069cc313
    • Debargha Mukherjee's avatar
      Make loop-restoration use 64x64 processing units · 7a5587a8
      Debargha Mukherjee authored
      Changes loop-restoration to use processing unit size that is
      64x64 for luma; for chroma the processing unit is coupled to
      64x64 support region for luma.
      Thus for chroma the processing unit size is 32x32 for 4:2:0,
      32x64 for 4:2:2 and 64x64 for 4:4:4, etc.
      
      While the Wiener filter output should not change with this patch,
      the sgr filter will change since the boundary pixel handling in
      sgr is internal within the filter.
      
      Change-Id: I65a9e2df88927a19445420ce400acb1fcf7afa93
      7a5587a8
  3. 04 Sep, 2017 2 commits
  4. 01 Sep, 2017 1 commit
    • Rupert Swarbrick's avatar
      Encode loop restoration coefficients per tile · 09b5b168
      Rupert Swarbrick authored
      This is a baby-step towards encoding the coefficients at the start of
      superblocks at the top-left of loop restoration tiles. Note that this
      patch causes us to reset "wiener_info" and "sgrproj_info" at each tile
      boundary, which will cause a performance drop.
      
      This is necessary because, in order for tiles to be processed in
      parallel, we cannot delta-encode coefficients across tile boundaries
      if the coefficients are signalled within tiles. We could probably do
      better than the current patch by, say, delta-encoding against previous
      frames.
      
      This patch also fixes up the costing in pickrst.c to match
      
      Change-Id: I5b8b91d63aaf49627cde40219c31c0ac776dfd38
      09b5b168
  5. 22 Aug, 2017 1 commit
  6. 31 Jul, 2017 1 commit
    • Yue Chen's avatar
      Move mode costs that can be updated inside a frame to MACROBLOCK · b23d00a0
      Yue Chen authored
      It is a refactoring patch, which aims to make the code ready for
      implementation of in-frame mode cost update in RDO.
      Also add mode cost update per sb row, but it won't affect coding
      results because cdf update in RDO is not there.
      Mode cost arrays are moved to MACROBLOCK because in multi-thread
      coding, threads share the same AV1_COMP.
      
      This patch does not have impact on coding results.
      
      Change-Id: I2e8f7d7d066b23ebfbfc998269023781f359a6ff
      b23d00a0
  7. 24 Jun, 2017 1 commit
  8. 15 Jun, 2017 1 commit
    • Urvang Joshi's avatar
      Remove 'rddiv' member from various structs. · 70006e46
      Urvang Joshi authored
      This was initialized from a const and never modified. But was still
      passed around and stored in multiple structs.
      
      Removed these 'rddiv' member variables and now RDOPT() and RDOPT_DBL()
      always use the const RDDIV_BITS directly.
      
      Change-Id: I1a8dfd2c8fa857d466ad1207b4f0dd6ec07eafb8
      70006e46
  9. 06 Jun, 2017 1 commit
    • Debargha Mukherjee's avatar
      Make loop-restoration compatible w/ frame_superres · 2dd982e4
      Debargha Mukherjee authored
      When frame_superres is on, loop-restoration should work
      on the size of the upscaled frame and not on the internal
      width and height in the common structure. This patch
      makes the necessary changes on the encoder and decoder
      side to enable that.
      
      Change-Id: I1d1c024ac6f95944169d90647b4c5a61354a5cc6
      2dd982e4
  10. 01 Jun, 2017 1 commit
  11. 21 May, 2017 1 commit
    • Debargha Mukherjee's avatar
      Cleans ups and refactoring · d48f573d
      Debargha Mukherjee authored
      Unifies the chroma and luma paths for estimation of the
      restoration parameters.
      
      No change in bit-stream or results.
      
      Change-Id: I9998c269cc00ded8b94d0e660b5441a4f9b94848
      d48f573d
  12. 19 May, 2017 1 commit
  13. 17 May, 2017 1 commit
  14. 15 May, 2017 1 commit
  15. 05 May, 2017 1 commit
    • Debargha Mukherjee's avatar
      Redo / refactor affine and rot-zoom least squares · 7ae7aeaf
      Debargha Mukherjee authored
      Use a simpler least-squares function for affine and rotzoom
      model estimation, instead of computing the pseudo inverse.
      Also refactors the code into a separate mathutils.h file.
      
      The SVD code is currently used only for estimation of the
      homography models which can be removed when we remove the
      homography models.
      
      Coding efficiency change is in noise range, with the small
      difference coming from numerical precision issues.
      
      Change-Id: I0a9eb79495911cea21a7945b397d596e22a2a186
      7ae7aeaf
  16. 03 May, 2017 1 commit
  17. 01 May, 2017 1 commit
    • Debargha Mukherjee's avatar
      Fix a bug in loop-restoration · 865a6f27
      Debargha Mukherjee authored
      Affected the switchable decision mechanism and
      effectively turned off block level switching, even
      though frame level switching was fine.
      
      Change-Id: Id0b19f91ecfae63964e73f1ba7bfeec70179da35
      865a6f27
  18. 17 Apr, 2017 1 commit
    • Jingning Han's avatar
      Singularity handling in Gaussian elimination · 041c67b9
      Jingning Han authored
      When the fwd Gaussian elimination process encounters diagonal
      element as zero value, the linear equation does not have unique
      solution. Return the linear solver state as unsolvable in such
      case. This resolves a floating point exception issue due to divided
      by zero in the loop restoration filter.
      
      BUG=aomedia:437
      
      Change-Id: I3c67525691a3003f9f470e8a0d5b4ee187cba963
      041c67b9
  19. 14 Apr, 2017 1 commit
  20. 12 Apr, 2017 1 commit
  21. 07 Apr, 2017 1 commit
  22. 14 Mar, 2017 1 commit
  23. 10 Mar, 2017 2 commits
    • David Barker's avatar
      Vectorize new highpass filter for loop-restoration · eed824ef
      David Barker authored
      Change-Id: Ibe5d4933f599456cb496f636de244694bc786a4c
      eed824ef
    • Debargha Mukherjee's avatar
      Replace one self guided filter with highpass · b7bb0976
      Debargha Mukherjee authored
      Adds an option controlled by a macro to replace one of
      the guided filters in the self-guided tool with a simple
      bandpass filtered version generated with a 3x3 kernel.
      By default the macro USE_HIGHPASS_IN_SGRPROJ is 0 (turned
      off), that defaults us to the dual self-guided filter.
      When the macro is turned on, the larger radius guided
      filter is replaced by a simpler filter that is much faster.
      
      Results (if USE_HIGHPASS_IN_SGRPROJ is on vs. off):
      lowres: performance drop by +0.14% (BDRATE)
      midres: performance drop by +0.27% (BDRATE)
      
      Further experiments on this variation of guided filters is
      pending.
      
      Change-Id: I7bbcfcad7ee266cd49a8dc6d96795a454feb1a94
      b7bb0976
  24. 09 Mar, 2017 3 commits
  25. 08 Mar, 2017 1 commit
    • David Barker's avatar
      Make encoder use vectorized self-guided filter · 506eb723
      David Barker authored
      By rearranging the code in restoration.c, we can allow the
      encoder to use the SSE4.1 version of the self-guided filter
      while picking the loop-restoration filter.
      
      This also helps us prepare for adding a highbitdepth SSE4.1
      version of the self-guided filter.
      
      No effect on encoder output, but gives an end-to-end speedup
      of 1-2%.
      
      Change-Id: Id17ba4a0963ddce9f70a7cae666e212e138d5f2c
      506eb723
  26. 06 Mar, 2017 2 commits
    • David Barker's avatar
      Vectorize self-guided filter · ce110cc5
      David Barker authored
      Add an SSE4.1 lowbd version of the self-guided filter for
      loop-restoration, and apply some optimizations to the C
      version.
      
      Approximate times per 128x128 / 256x256 tile on the machine
      this was developed on:
      Previous C:  620us / 2800us
      Optimized C: 500us / 2200us ( 24% /  27% faster)
      SSE4.1:      147us / 600us  (320% / 370% faster)
      
      Change-Id: I23ff5a5482a191aeb06f9d1f767a9f036bb357fe
      ce110cc5
    • Debargha Mukherjee's avatar
      Remove unnecessary filtering and buffer copies · 00c54330
      Debargha Mukherjee authored
      Removes redundant deblocking filtering operations and buffer copies,
      during search for the restoration filters.
      Also, makes the order of cdef and loop-restoration consistent.
      
      BUG=AOMEDIA:373
      
      Change-Id: I571e331f9dfadfd2c1494f279e50510ec212ea29
      00c54330
  27. 14 Feb, 2017 1 commit
  28. 13 Feb, 2017 1 commit
  29. 19 Jan, 2017 1 commit
  30. 18 Jan, 2017 1 commit
  31. 09 Jan, 2017 2 commits
  32. 07 Jan, 2017 2 commits
    • David Barker's avatar
      Optimize Wiener filter selection · 33f3bfde
      David Barker authored
      * Change the behaviour of search_wiener at borders to match
        the behaviour of the Wiener filter itself
      * Reorder the calculation in compute_stats, saving ~5% of
        encode time at low bitrates (tested on bus_cif.y4m at 200kbps)
      
      Change-Id: I5f649d77fd66584451aaf37697ce9c9af69524e4
      33f3bfde
    • Debargha Mukherjee's avatar
      A mismatch fix in loop restoration · 09ad6d85
      Debargha Mukherjee authored
      Change-Id: Icfc4645ff97d4fd6849f149f4c5296a53c204cf4
      09ad6d85