1. 31 Oct, 2018 17 commits
  2. 20 Oct, 2018 2 commits
    • David Flynn's avatar
      tools/cfg-gen: process all ctc yaml files simultaneously · 97762fb5
      David Flynn authored
      gen-cfg.sh processes each ctc yaml file in turn, which was advantageous
      when the ctc yaml files contained multiple conflicting definitions for
      a test condition (eg, one defines *-lossy-attrs for colour, and another
      defines *-lossy-attrs for reflectance).  In such cases, merging the
      two yaml files together is not possible.
      However, this does prevent having a single yaml file that modifies
      the parameters for multiple test conditions. (Eg, adding sequences
      to all ctc test conditions).
    • David Flynn's avatar
      tools/cfg-gen: apply extra config files after hard coded defaults · 6139d1f8
      David Flynn authored
      In order to allow extra config files to override those hard-coded
      in the gen-cfg.sh script, they need to be specified at the end.
  3. 05 Sep, 2018 3 commits
  4. 03 Sep, 2018 14 commits
  5. 29 Aug, 2018 4 commits
    • Khaled Mammou's avatar
      cfg: automatic derivation of dist2 series · 78a60bce
      Khaled Mammou authored and David Flynn's avatar David Flynn committed
      This commit introduces automatic derivation of dist2 values by the
      frontend.  Automatic derivation occurs if fewer than levelOfDetailCount
      values are provided.  No syntax changes are made.
      cfg update:
       - use single dist2 values based on existing config
       - ford: 16383 -> 16384
    • Khaled Mammou's avatar
      attr: remove no longer used PCCPredictor::levelOfDetailIndex · 764d40d4
      Khaled Mammou authored and David Flynn's avatar David Flynn committed
      This functionality is unnecessary given a uniform quant_step size.
    • Khaled Mammou's avatar
      attr: use single quant_step_size_* for all LoDs in lift/pred · 51f21bcb
      Khaled Mammou authored and David Flynn's avatar David Flynn committed
      This commit removes the ability to vary quant_step_size on a per-LoD
      basis.  This greatly simplifies the codec configuration, especially
      when used in conjunction with dist2 derivation.
      cfg update:
       - rename quantizationSteps* -> quantizationStep*
       - update quantizationStep* to single values:
         - lossless-geom lossy-attrs:        8, 16, 32, 64, 128, 256 [8bit]
         - lossless-geom nearlossless-attrs: no change
         - lossless-geom lossless-attrs:     no change
         - lossy-geom    lossy-attrs:        8                       [8bit]
       => 16-bit cat3 sequences are have 8-bit qs values multipled by 255
    • David Flynn's avatar
      attr: consistently declare computeReflectancePredictionWeights · 983f6388
      David Flynn authored
      This commit maintains consistency with other static AttributeEncoder
      and AttributeDecoder methods.