1. 15 Jan, 2015 1 commit
    • Frank Galligan's avatar
      Add Neon intrinsics for vp9_avg_8x8_neon · 6e7e1cf3
      Frank Galligan authored
      On Nexus 7 speed -5, -6, -7, and -8 saw about a 1% increase
      in perf for 480p. Speeds -5, -6, -7, and -8 saw about a 1.5%
      increase in perf for 720p.
      Tested on Nexus 7, built with ndk r10d, gcc 4.9.
      Change-Id: Ibf17ebfd952a6aec941719bd8306df8ec4574bee
  2. 14 Jan, 2015 4 commits
    • Yunqing Wang's avatar
      Align thread data in vp9_ethread · 99b99831
      Yunqing Wang authored
      On some platforms, such as 32bit Windows and 32bit Mac, the allocated
      memory isn't aligned automatically. The thread data is aligned to
      ensure the correct access in SIMD code.
      Change-Id: I1108c145fe982ddbd3d9324952758297120e4806
    • Yaowu Xu's avatar
      Add encoder control for setting color space · e94b415c
      Yaowu Xu authored
      This commit adds encoder side control for vp9 to set color space info
      in the output compressed bitstream.
      It also amends the "vp9_encoder_params_get_to_decoder" test to verify
      the correct color space information is passed from the encoder end to
      decoder end.
      Change-Id: Ibf5fba2edcb2a8dc37557f6fae5c7816efa52650
    • Frank Galligan's avatar
      Add 64x64 sub_pel_variance Neon function · ec1d8387
      Frank Galligan authored
      On Nexus 7 speed -5, -6, -7, and -8 saw about a 15% increase
      in perf for 480p. Speeds -5, -6, -7, and -8 saw about a 10%
      increase in perf for 720p.
      Tested on Nexus 7, built with ndk r10d, gcc 4.9.
      Change-Id: I2fa5315845e3021c9a6e2ea47e52e68b398d8334
    • Frank Galligan's avatar
      Switch remaining Neon variance functions to shifts · 588f74f8
      Frank Galligan authored
      Saves 5 instructions on 8x8 and 16x16 and 8 instructions
      on 32x32, when compiled with 4.9.
      Change-Id: Id3da613a36a9d27d8c5169c59ba45d247c920c6c
  3. 13 Jan, 2015 3 commits
    • Minghai Shang's avatar
      [twopass temporal svc] Fix decoding error on seek. · a14415d1
      Minghai Shang authored
      Don't put small empty frame in front of a key frame. We will
      put key frame flag in webm container if there's a visible key
      frame. But there will be decoding error when we seek to here
      if we put the small empty frame, which will be inter frame,
      in front of it.
      Change-Id: Id50c2c1fd31da0405ff6faa7375cc2f49c55402d
    • Yaowu Xu's avatar
      Enable decoder to pass through color space info · 6b223fcb
      Yaowu Xu authored
      This commit added a field to vpx_image_t for indicating color space,
      the field is also added to YUV_BUFFER_CONFIG. This allows the color
      space information pass through the decoder from input stream to the
      output buffer.
      The commit also updated compare_img() function with added verification
      of matching color space to ensure the color space information to be
      correctly passed from encode to decoder in compressed vp9 streams.
      Change-Id: I412776ec83defd8a09d76759aeb057b8fa690371
    • Frank Galligan's avatar
      Add 64x variance Neon functions · 74d40cd5
      Frank Galligan authored
      Add optimized Neon functions of:
      On Nexus 7 speed -5 and -6 saw about a 4% increase in perf.
      Speeds -7 and -8 saw about a 6% increase in perf.
      Tested on Nexus 7, built with ndk r10d, gcc 4.9.
      Change-Id: I5a81f13c9897eb927fa39662530f5524a0f768fa
  4. 09 Jan, 2015 2 commits
    • Yaowu Xu's avatar
      Added plumbing for setting color space · ce52b0f8
      Yaowu Xu authored
      Change-Id: If64052cc6e404abc8a64a889f42930d14fad21d3
    • Yaowu Xu's avatar
      Fix comments and color format · ecbca31a
      Yaowu Xu authored
      Replaced "color space" with "color format" in comments where color
      sampling format is concerned, so to differentiate from the concept
      defined in COLOR_SPACE.
      Change-Id: I8c935034c166b24307a99352dab1686531276bb8
  5. 08 Jan, 2015 5 commits
  6. 07 Jan, 2015 12 commits
  7. 06 Jan, 2015 4 commits
    • Yaowu Xu's avatar
      Use -1 consistently as invalid buffer idx · 61c5e94e
      Yaowu Xu authored
      Instead of mixed use of both -1 and INT_MAX.
      This also fixes a vp9 fuzzing test failure.
      Change-Id: I950ea94b44ec7cdb5232773bee30b104e342f52a
    • Deb Mukherjee's avatar
      Enable coefficient range checking for 10-/12-bit · 0ce2a27e
      Deb Mukherjee authored
      Also fixes a broken build with --enable-coefficient-range-checking
      configuration option.
      Change-Id: Icc536f53088e8cec59dfb8f635668555fdb9125e
    • Yaowu Xu's avatar
      Properly validate data size · 9c061ef5
      Yaowu Xu authored
      With "show_existing_frame" frames:
      Minimum data size for profile 0 and 1 is 1 byte (8bits)
      Minimum data size for profile 2 and 3 is 2 bytes (9bits)
      Minimum data size is 8 bytes.
      This resolves the VP9 failure in fuzzing test build #56.
      Change-Id: I146d9d37688f535dd68d24aacc76d464ccffdf04
    • Yaowu Xu's avatar
      Fix compiler warnigns for msvc2013 · 364b92dc
      Yaowu Xu authored
      Change-Id: I1e32bf8f6872a6fb7e9cabe86483e94805e2f790
  8. 05 Jan, 2015 1 commit
    • JackyChen's avatar
      Adopt weighted averaging in MFQE. · fe23539d
      JackyChen authored
      By using weighted averaging in the calculation of the frames to be
      displayed, we get an average gain of more than 1 db for key frames
      whose base qp are 20 higher than non-key frames.
      Change-Id: I7bcb2e7b9c6420ea3f73f33204d18b072dffd17c
  9. 04 Jan, 2015 1 commit
    • Jingning Han's avatar
      Fix denoised video output function · 21c03061
      Jingning Han authored
      This commit fixes the buffer alignment control in denoised video
      output function. The encoder is now able to properly store the
      denoised input video into provided file when enabled.
      Change-Id: I258e272c8d4a9b52592e16d6d09976c6f5c21728
  10. 02 Jan, 2015 2 commits
  11. 30 Dec, 2014 2 commits
  12. 25 Dec, 2014 1 commit
    • Jingning Han's avatar
      Enable sub8x8 inter block search for RTC coding mode · dad89d5c
      Jingning Han authored
      This commit enables sub8x8 inter block coding for RTC mode. The
      use of sub8x8 blocks can be turned on by allowing
      choose_partitioning function to select 4x4/4x8/8x4 block sizes.
      Change-Id: Ifbf1fb3888fe4c094fc85158ac3aa89867d8494a
  13. 23 Dec, 2014 2 commits