]> granicus.if.org Git - libvpx/commit
projects / libvpx / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 0c9e2e9) | patch
optimize forward 16x16 DCT for accuracy
authorYaowu Xu <yaowu@google.com>
Sat, 23 Feb 2013 00:18:08 +0000 (16:18 -0800)
committerYaowu Xu <yaowu@google.com>
Mon, 25 Feb 2013 15:47:27 +0000 (07:47 -0800)
commit499fe05dc030b31d15c2ad9717e6bcee839f7d12
tree0a1940649fca62295840ea1ccdecb40d95d743b8tree | snapshot
parent0c9e2e9a1ddca6e120c4afa7f225968a9ab989bfcommit | diff
optimize forward 16x16 DCT for accuracy

This commit added pre/post scaling for first half of fDCT16x16 to
reduce error, by simulation of 100,000 blocks for random inputs,
the average sse reduced from 2.1/block to 0.0498/block.

also enabled tests for 16x16 fDCT and iDCT

Change-Id: Id2a95f0464c6dd4118797d456237ae90274c0f02
test/dct16x16_test.cc diff | blob | history
test/test.mk diff | blob | history
vp9/encoder/vp9_dct.c diff | blob | history
Unnamed repository; edit this file 'description' to name the repository.