1. 27 Jun, 2012 1 commit
    • Adrian Grange's avatar
      Added Prediction Filter to Mode Selection · bbc926dc
      Adrian Grange authored
      Added the ability to optionally filter the prediction data
      when inter modes are selected (excludes SPLITMV, for now).
      
      The mode selection loop considers both the filtered and
      non-filtered prediction data when choosing mode. The filter
      can be turned on/off at the frame-level, or signaled for
      each MB.
      
      Change-Id: I1b783c71d95a361ab36c761b07e8a6b06bc36822
      bbc926dc
  2. 26 Jun, 2012 1 commit
    • Deb Mukherjee's avatar
      Continued adaptive entropy coding · f3dff402
      Deb Mukherjee authored
      Incorporates mv_ref, mbsplit and second_mv into the adaptive
      entropy framework. The mv_ref framework has been modified from
      before.
      
      Adds some clean-ups and fixes.
      
      Results with the adaptive entropy experiment are currently up by
      +1.93% on derf; +2.33% std-hd and +1.87% yt-hd.
      
      Fixed a nasty intermittent bug.
      
      Change-Id: I4b1ac9f9483b48432597595195bfec05f31d1e39
      f3dff402
  3. 19 Jun, 2012 1 commit
  4. 15 Jun, 2012 1 commit
    • Deb Mukherjee's avatar
      Adaptive entropy coding of coefficients, modes, mv. · 1fe85a35
      Deb Mukherjee authored
      This patch incorporates adaptive entropy coding of coefficient tokens,
      and mode/mv information based on distributions encountered in a frame.
      Specifically, there is an initial forward update to the probabilities
      in the bitstream as before for coding the symbols in the frame, however
      at the end of decoding each frame, the forward update to the
      probabilities is reverted and instead the probabilities are updated
      towards the actual distributions encountered within the frame.
      The amount of update is weighted by the number of hits within each
      context.
      
      Results on derf/hd/std-hd are all up by 1.6%.
      
      On derf, the most of the gains come from coefficients, however for the
      hd and std-hd sets, the most of the gains come from the mode/mv
      information updates.
      
      Change-Id: I708c0e11fdacafee04940fe7ae159ba6844005fd
      1fe85a35
  5. 12 Jun, 2012 1 commit
    • Yaowu Xu's avatar
      Added the ability to accumulate coef stats across encodings · bb25083d
      Yaowu Xu authored
      This commit added the ability to accumulate the coef stats across
      different encodings using an intermediate binary stats files. The
      accumulation happens only the binary stats file exists in current
      directory. The encoder needs to be built with "ENTROPY_STATS" to
      allow the output. The commit also fixed a few formating issues in
      output stats file.
      
      Change-Id: Ib1a41180aa554845cf51e4421a230b128a3a82b4
      bb25083d
  6. 08 Jun, 2012 1 commit
    • Paul Wilkins's avatar
      Key frame and GF sizing changes. · 0cb4985b
      Paul Wilkins authored
      Changes to the equations for kf and gf minQ and to
      the boost calculations for kg and gf
      
      Change-Id: I312031c910e6a575334f49075c32f49a8dfff239
      0cb4985b
  7. 15 May, 2012 1 commit
  8. 11 May, 2012 1 commit
    • Paul Wilkins's avatar
      First pass overhaul preparatory change. · 35358320
      Paul Wilkins authored
      This is the first patch in a series of changes to the first
      pass code. (Broken down for ease of testing/merging/review).
      
      This patch introduces a new stats element "sr_coded_error".
      This is the coded error recorded vs the second reference
      frame (which is updated such that it lags by at least one frame).
      No use is made of the new structure in this change so this patch
      should have no material effect.
      
      Removed some ifdefs and deprecated code (#if NEW_BOOST).
      Removed twopass.gf_decay_rate (not used any more)
      
      Change-Id: I1be672a73017f7c13fd50fb4f99236aa2ed30916
      35358320
  9. 08 May, 2012 1 commit
    • Yaowu Xu's avatar
      a number of fixes to entropy stats collection · 54cf1d9a
      Yaowu Xu authored
      1. block types
      There are only three types of blocks for 8x8 transformed MBs, i.e. Y
      block with DC does not exist for 8x8 transformed MBs as all MB using
      8x8 transform have 2nd order haar transform. This commit introduced
      a new macro BLOCK_TYPES_8X8 to reflect such fact.
      
      2. context counters
      This commit also fixed the mixed of context_counters between 4x4 and
      8x8 transformed MBs. The mixed use of the counters leads me to think
      the existing the context probabilities were not properly generated
      from 8x8 transformed MBs.
      
      3. redundant collecting in recoding
      The commit also corrected the code that accumulates entropy stats by
      making sure stats only collected for final packing, not during the
      recode loop
      
      Change-Id: I029f09f8f60bd0c3240cc392ff5c6d05435e322c
      54cf1d9a
  10. 30 Apr, 2012 1 commit
  11. 26 Apr, 2012 1 commit
    • Deb Mukherjee's avatar
      Adds search option for best interpolation filter. · acdda50a
      Deb Mukherjee authored
      Adds a speed feature to conduct a brute force search among a set of
      available interpolation filters for the best one in an RD sense.
      
      There is a gain of 0.4% on derf, 1.0% on Std-HD.
      
      Patch 2: A macro added to determine if the encoder state is reset
      for each new filter tried.
      
      Patch 3: rebase, also fixes a bug (decodframe.c) introduced by a
      couple of missing function pointer assignements.
      
      Patch 4: rebase.
      
      Change-Id: Ic9ccca9d8c35c6af557449ae867391a2f996cc29
      acdda50a
  12. 25 Apr, 2012 1 commit
    • Yaowu Xu's avatar
      Merge UVINTRA experiment · c1814d26
      Yaowu Xu authored
      The commit merges the UVINTRA experiment and removed the related
      macros. The overall effect of the experiment is a small gain (.1%
      on derf)
      
      Change-Id: Ia34b3312fb9b5b34c9ba111bf0fa78c6f78ac80b
      c1814d26
  13. 18 Apr, 2012 1 commit
  14. 12 Apr, 2012 1 commit
    • Deb Mukherjee's avatar
      Turning off interpolation filter selection · 237718dc
      Deb Mukherjee authored
      Turning off the interpolation filter selection based on edge
      proportion. This heuristics has not been working as well as
      expected and I have started a more rigorous investigation into
      this. We can turn this off for now since it is unnecessarily
      slowing things down.
      
      Rebase.
      
      Change-Id: Ic5958b2b3a35ec2d8eb73b6d81617ca8fbe07e74
      237718dc
  15. 11 Apr, 2012 5 commits
    • Paul Wilkins's avatar
      Delete unused function. · d6ac213c
      Paul Wilkins authored
      Deleted check_gf_quality().
      
      Change-Id: If75fbd84accb1f6471ad6ea6ff2b65ae99976f5d
      d6ac213c
    • Paul Wilkins's avatar
      Refactoring of encode loop and bitstream packing · 13c6d1a8
      Paul Wilkins authored
      Some code re-factored / moved to allow the main
      pack operation inside the recode loop so  that the
      size estimate is accurate.
      
      Deletion of some redundant code relating to one pass.
      
      Aproximate improvement over March 27 code base:
      Derf 0.0%, YT 0.5%, YThd 0.3% Std_hd 0.25%
      
      Change-Id: Id2d071794ab44f0b52935f6fcdb5733d09a6bb86
      13c6d1a8
    • Paul Wilkins's avatar
      Changes to costing of skip. · f2ec452f
      Paul Wilkins authored
      Update the costing of skip in the recode loop and rd code.
      
      Change-Id: I2e5ebbd7ddf201212b32441321e12626cd0423e9
      f2ec452f
    • Paul Wilkins's avatar
      T8x8 zbin and rate control changes. · a3392d57
      Paul Wilkins authored
      Some adjustments to zbin for t8x8.
      Changes to rules for sizing forced key frames.
      Some extra stats output in tmp.stt.
      
      Approximate  gain on YT-hd set 0.5%
      
      There are still issues in sizing key frames and gf/arf frames
      when the image is largely static. These in part relate to
      problems with cost estimates in the recode loop.
      
      Change-Id: I6f0159dc8a8faeab4115a19c668d442491619a68
      a3392d57
    • Adrian Grange's avatar
      Superblock encoding order · 9daf3154
      Adrian Grange authored
      This is the first patch to add superblock (32x32) coding
      order capabilities. It does not yet do any mode selection
      at the SB level, that will follow in a further patch.
      
      This patch encodes rows of SBs rather than
      MBs, each SB contains 2x2 MBs.
      
      Two intra prediction modes have been disabled since they
      require reconstructed data for the above-right MB which
      may not have been encoded yet (e.g. for the bottom right
      MB in each SB).
      
      Results on the one test clip I have tried (720p GIPS clip)
      suggest that it is somewhere around 0.2dB worse than the
      baseline version, so there may be bugs.
      
      It has been tested with no experiments enabled and with
      the following 3 experiments enabled:
        --enable-enhanced_interp
        --enable-high_precision_mv
        --enable-sixteenth_subpel_uv
      in each case the decode buffer matches the recon buffer
      (using "cmp" to compare the dumped/decoded frames).
      Note: Testing these experiments individually created
      errors.
      
      Some problems were found with other experiments but it
      is unclear what state these experiments are in:
        --enable-comp_intra_pred
        --enable-newentropy
        --enable-uvintra
      
      This code has not been extensively tested yet, so there
      is every likelihood that further bugs remain. I also
      intend to do some code cleanup & refactoring in tandem
      with the next patch that adds the 32x32 modes.
      
      Change-Id: I1eba7f740a70b3510df58db53464535ef881b4d9
      9daf3154
  16. 21 Mar, 2012 3 commits
    • Paul Wilkins's avatar
      Only support improved quant · c88d335f
      Paul Wilkins authored
      Deprecate fast quant and strict_quant code.
      Small effect on quality as fast was used in first pass but the
      effect is basically neutral across the derf set.
      
      The rationale here is to reduce the number of code paths for
      now to make experimentation easier. Optimized and fast code
      options can be re-introduced later along with other  encode
      speed options.
      
      Change-Id: Ia30c5daf3dbc52e72c83b277a1d281e3c934cdad
      c88d335f
    • Deb Mukherjee's avatar
      Making subpel filters switchable at frame level · 475d5d56
      Deb Mukherjee authored
      Various refactoring to make the subpel motion compensation
      filters switchable by a frame level field.
      Two types of 8-tap filters are supported in addition to the existing
      bilinar and sixtap filters. One is the default 8-tap and the
      other has a sharper cut-off for use with frames with substantial
      edge content.
      
      Patch 2: Added a preliminary strategy for filter selection based on
      edginess detecton. Also includes some filter changes.
      
      Change-Id: I866085bda5ae143cfdf2ec88157feaabdf7bd63a
      475d5d56
    • Deb Mukherjee's avatar
      Adding contextual coding of mb_skip_coeff flag. · 57d95347
      Deb Mukherjee authored
      Using contextual coding of the mkb_skip_coeff flag using the
      values of this flag from the left and above. There is a small
      improvement of about 0.15% on Derf:
      http://www.corp.google.com/~debargha/vp8_results/mbskipcontext.html
      
      Refactored to use pred_common.c by adding a new context type.
      
      Results on HD set (about 0.66% improvement):
      http://www.corp.google.com/~debargha/vp8_results/mbskipcontext_hd.html
      
      Incliding missing refactoring to use the pred_common utilities.
      
      Change-Id: I95373382d429b5a59610d77f69a0fea2be628278
      57d95347
  17. 05 Mar, 2012 1 commit
  18. 01 Mar, 2012 1 commit
  19. 29 Feb, 2012 1 commit
  20. 28 Feb, 2012 2 commits
    • Deb Mukherjee's avatar
      Initial refactoring of high_precision mv code. · 3e1cad9c
      Deb Mukherjee authored
      This is the first patch for refactoring of the code related to
      high-precision mv, so that 1/4 and 1/8 pel motion vectors can
      co-exist in the same bit-stream by use of a frame level flag.
      The current patch works fine for only use of 1/4th and
      only use of 1/8th pel mv, but there are some issues with the
      mode switching in between. Subsequent patches on this change Id
      will fix the remaining issues.
      
      Patch 2: Adds fixes to make sure that multiple mv precisions can
      co-exist in the bit-stream. Frame level switching has been tested
      to work correctly.
      
      Patch 3: Fixes lines exceeding 80 char
      
      Patch 4:
      http://www.corp.google.com/~debargha/vp8_results/enhinterp.html
      
      Results on derf after ssse3 bugfix, compared to everything
      enabled but the 8-tap, 1/8-subpel and 1/16-subpel uv. Overall the
      gains are about 3% now. Hopefully there are no more bugs lingering.
      Apparently the sse3 bug affected the quartel subpel results more than
      the eighth pel ones (which is understandabale because one bad predictor
      due to the bug, matters less if there are a lot more subpel options
      available as in the 1/8 subpel case).
      The results in the 4th column correspond to the current settings.
      The first two columns correspond to two settings of adaptive switching
      of the 1/4 or 1/8 subpel mode based on initial Q estimate. These
      do not work as good as just using 1/8 all the time yet.
      
      Change-Id: I3ef392ad338329f4d68a85257a49f2b14f3af472
      3e1cad9c
    • Paul Wilkins's avatar
      Code Simplification · b6f02c85
      Paul Wilkins authored
      Removal of code relating to token partitioning
      
      Change-Id: Iaf3c88d6758639a55bd92c3be5c51e6bed407a3c
      b6f02c85
  21. 27 Feb, 2012 1 commit
    • Paul Wilkins's avatar
      Corrected spelling · b00ed02a
      Paul Wilkins authored
      Apparently the correct spelling of segement is segment !
      
      Change-Id: I88593ee0523f251b3a96794c6166ef8c7898a029
      b00ed02a
  22. 26 Feb, 2012 1 commit
    • Paul Wilkins's avatar
      Code Cleanup. · 646e6221
      Paul Wilkins authored
      Removal of error_resilient_mode features.
      The interface has been left in place but does nothing.
      
      Change-Id: I2407863bd0d3c98407354507423ca48d29f63b17
      646e6221
  23. 25 Feb, 2012 1 commit
    • Paul Wilkins's avatar
      Removal of temporal re sampling code. · 80b873e3
      Paul Wilkins authored
      For now the interface elements have been left in place
      to make sure existing parameter files work but parameters
      relating to drop frame wont do anything.
      
      Change-Id: I579ee614726387381c546845dac4bc03c74c6a07
      80b873e3
  24. 24 Feb, 2012 2 commits
  25. 23 Feb, 2012 1 commit
    • Deb Mukherjee's avatar
      Supporting high precision 1/8-pel motion vectors · 18e90d74
      Deb Mukherjee authored
      This is the initial patch for supporting 1/8th pel
      motion. Currently if we configure with enable-high-precision-mv,
      all motion vectors would default to 1/8 pel. Encode and
      decode syncs fine with the current code. In the next phase
      the code will be refactored so that we can choose the 1/8
      pel mode adaptively at a frame/segment/mb level.
      
      Derf results:
      http://www.corp.google.com/~debargha/vp8_results/enhinterp_hpmv.html
      (about 0.83% better than 8-tap interpoaltion)
      
      Patch 3: Rebased. Also adding 1/16th pel interpolation for U and V
      
      Patch 4: HD results.
      http://www.corp.google.com/~debargha/vp8_results/enhinterp_hd_hpmv.html
      Seems impressive (unless I am doing something wrong).
      
      Patch 5: Added mmx/sse for bilateral filtering, as well as enforced
      use of c-versions of subpel filters with 8-taps and 1/16th pel;
      Also redesigned the 8-tap filters to reduce the cut-off in order to
      introduce a denoising effect. There is a new configure option
      sixteenth-subpel-uv which will use 1/16 th pel interpolation for
      uv, if the motion vectors have 1/8 pel accuracy.
      
      With the fixes the results are promising on the derf set. The enhanced
      interpolation option with 8-taps alone gives 3% improvement over thei
      derf set:
      http://www.corp.google.com/~debargha/vp8_results/enhinterpn.html
      
      Results on high precision mv and on the hd set are to follow.
      
      Patch 6: Adding a missing condition for CONFIG_SIXTEENTH_SUBPEL_UV in
      vp8/common/x86/x86_systemdependent.c
      
      Patch 7: Cleaning up various debug messages.
      
      Patch 8: Merge conflict
      
      Change-Id: I5b1d844457aefd7414a9e4e0e06c6ed38fd8cc04
      18e90d74
  26. 17 Feb, 2012 1 commit
    • Paul Wilkins's avatar
      Code base simplification. · 4cfb8ed4
      Paul Wilkins authored
      Removal of most code to do with 1 pass.
      Removal of cyclic refresh code.
      
      Change-Id: I74971082bc19dd76e795d4d2e781a0424cec5c8c
      4cfb8ed4
  27. 16 Feb, 2012 2 commits
    • Paul Wilkins's avatar
      Code simplification · 79d330d7
      Paul Wilkins authored
      Removal of the pickinter.c and .h files and calls to this
      code.
      
      Removal of some code relating to real time and one pass
      settings  though there is more to be done in this regard.
      
      However,  vp8_set_speed_features() now
      only supports modes 0 and 1 and speeds up to 3
      so rd should always be set.
      
      Change-Id: I62c0c1b6154ab499785baef310536080e87bc4d8
      79d330d7
    • Yaowu Xu's avatar
      moved scaling from dequantization to inverse transform for T8x8 · 454c7abc
      Yaowu Xu authored
      Previously, the scaling related to extended quantize range happens in
      dequantization stage, which implies the coefficients form forward
      transform are in different scale(4x) from dequantization coefficients
      This worked fine when there was not distortion computation done based
      on 8x8 transform, but it completely wracked the distortion estimation
      based on transform coefficients and dequantized transform coefficients
      introduced in commit f64725a0 for macroblocks using 8x8 transform.
      This commit fixed the issue by moving the scaling into the stage of
      inverse 8x8 transform.
      
      TODO: Test&Verify the transform/quantization pipeline accuracy.
      
      Change-Id: Iff77b36a965c2a6b247e59b9c59df93eba5d60e2
      454c7abc
  28. 14 Feb, 2012 1 commit
  29. 10 Feb, 2012 3 commits
    • Paul Wilkins's avatar
      Removal of threading code. · 2615ca5d
      Paul Wilkins authored
      For the experimental branch we are trying to slim the codebase
      down removing features such as threading for now which complicate
      the process of development and testing.
      
      Change-Id: I657c0246aef4d1fa8c8ffc6a1adfeee45bce8e24
      2615ca5d
    • Ronald S. Bultje's avatar
      Improved coding using 8x8 transform · f64725a0
      Ronald S. Bultje authored
      In summary, this commit encompasses a series of changes in attempt to
      improve the 8x8 transform based coding to help overall compression
      quality, please refer to the detailed commit history below for what
      are the rationale underly the series of changes:
      
      a. A frame level flag to indicate if 8x8 transform is used at all.
      b. 8x8 transform is not used for key frames and small image size.
      c. On inter coded frame, macroblocks using modes B_PRED, SPLIT_MV
      and I8X8_PRED are forced to using 4x4 transform based coding, the
      rest uses 8x8 transform based coding.
      d. Encoder and decoder has the same assumption on the relationship
      between prediction modes and transform size, therefore no signaling
      is encoded in bitstream.
      e. Mode decision process now calculate the rate and distortion scores
      using their respective transforms.
      
      Overall test results:
      1. HD set
      http://www.corp.google.com/~yaowu/no_crawl/t8x8/HD_t8x8_20120206.html
      (avg psnr: 3.09% glb psnr: 3.22%, ssim: 3.90%)
      2. Cif set:
      http://www.corp.google.com/~yaowu/no_crawl/t8x8/cif_t8x8_20120206.html
      (avg psnr: -0.03%, glb psnr: -0.02%, ssim: -0.04%)
      It should be noted here, as 8x8 transform coding itself is disabled
      for cif size clips, the 0.03% loss is purely from the 1 bit/frame
      flag overhead on if 8x8 transform is used or not for the frame.
      
      ---patch history for future reference---
      Patch 1:
      this commit tries to select transform size based on macroblock
      prediction mode. If the size of a prediction mode is 16x16, then
      the macroblock is forced to use 8x8 transform. If the prediction
      mode is B_PRED, SPLITMV or I8X8_PRED, then the macroblock is forced
      to use 4x4 transform. Tests on the following HD clips showed mixed
      results: (all hd clips only used first 100 frames in the test)
      
      http://www.corp.google.com/~yaowu/no_crawl/t8x8/hdmodebased8x8.html
      http://www.corp.google.com/~yaowu/no_crawl/t8x8/hdmodebased8x8_log.html
      
      while the results are mixed and overall negative, it is interesting to
      see 8x8 helped a few of the clips.
      
      Patch 2:
      this patch tries to hard-wire selection of transform size based on
      prediction modes without using segmentation to signal the transform size.
      encoder and decoder both takes the same assumption that all macroblocks
      use 8x8 transform except when prediciton mode is B_PRED, I8X8_PRED or
      SPLITMV. Test results are as follows:
      
      http://www.corp.google.com/~yaowu/no_crawl/t8x8/cifmodebase8x8_0125.html
      http://www.corp.google.com/~yaowu/no_crawl/t8x8/hdmodebased8x8_0125log.html
      
      Interestingly, by removing the overhead or coding the segmentation, the
      results on this limited HD set have turn positive on average.
      
      Patch 3:
      this patch disabled the usage of 8x8 transform on key frames, and kept the
      logic from patch 2 for inter frames only. test results on HD set turned
      decidedly positive with 8x8 transform enabled on inter frame with 16x16
      prediction modes: (avg psnr: .81% glb psnr: .82 ssim: .55%)
      
      http://www.corp.google.com/~yaowu/no_crawl/t8x8/hdintermode8x8_0125.html
      results on cif set still negative overall
      
      Patch 4:
      continued from last patch, but now in mode decision process, the rate and
      distortion estimates are computed based on 8x8 transform results for MBs
      with modes associated with 8x8 transform. This patch also fixed a problem
      related to segment based eob coding when 8x8 transform is used. The patch
      significantly improved the results on HD clips:
      
      http://www.corp.google.com/~yaowu/no_crawl/t8x8/hd8x8RDintermode.html
      (avg psnr: 2.70% glb psnr: 2.76% ssim: 3.34%)
      results on cif also improved, though they are still negative compared to
      baseline that uses 4x4 transform only:
      http://www.corp.google.com/~yaowu/no_crawl/t8x8/cif8x8RDintermode.html
      (avg psnr: -.78% glb psnr: -.86% ssim: -.19%)
      
      Patch 5:
      This patch does 3 things:
      a. a bunch of decoder bug fixes, encodings and decodings were verified
      to have matched recon buffer on a number of encodes on cif size mobile and
      hd version of _pedestrian.
      b. the patch further improved the rate distortion calculation of MBS that
      use 8x8 transform. This provided some further gain on compression.
      c. the patch also got the experimental work SEG_LVL_EOB to work with 8x8
      transformed macroblock, test results indicates it improves the cif set
      but hurt the HD set slightly.
      
      Tests results on HD clips:
      http://www.corp.google.com/~yaowu/no_crawl/t8x8/HD_t8x8_20120201.html
      (avg psnr: 3.19% glb psnr: 3.30% ssim: 3.93%)
      
      Test results on cif clips:
      http://www.corp.google.com/~yaowu/no_crawl/t8x8/cif_t8x8_20120201.html
      (avg psnr: -.47% glb psnr: -.51% ssim: +.28%)
      
      Patch 6:
      Added a frame level flag to indicate if 8x8 transform is allowed at all.
      temporarily the decision is based on frame size, can be optimized later
      one. This get the cif results to basically unchanged, with one bit per
      frame overhead on both cif and hd clips.
      
      Patch 8:
      Rebase and Merge to head by PGW.
      Fixed some suspect 4s that look like hey should be 64s in regard
      to segmented EOB. Perhaps #defines would be bette.
      Bulit and tested without T8x8 enabled and produces unchanged
      output.
      
      Patch 9:
      Corrected misalligned code/decode of "txfm_mode" bit.
      Limited testing for correct encode and decode with
      T8x8 configured on derf clips.
      
      Change-Id: I156e1405d25f81579d579dff8ab9af53944ec49c
      f64725a0
    • Ronald S. Bultje's avatar
      Merge dualpred (compound prediction) experiment. · 29e4d7e8
      Ronald S. Bultje authored
      Change-Id: Ieaaa07c50eae41118596197f6a4d848135946e41
      29e4d7e8