diff options
| author | Jingning Han <jingning@google.com> | 2014-06-25 16:53:07 -0700 |
|---|---|---|
| committer | Jingning Han <jingning@google.com> | 2014-06-26 16:00:42 -0700 |
| commit | 5a3e3c6d3fa308066e2cef1f8cbc407cd540c176 (patch) | |
| tree | 9001ca15c0cc30288a4d837b23207440fd859a9d /vp9/encoder/vp9_encodeframe.c | |
| parent | 73eeb3beffefb5238071988fbea94cba9c2b558b (diff) | |
| download | libvpx-5a3e3c6d3fa308066e2cef1f8cbc407cd540c176.tar libvpx-5a3e3c6d3fa308066e2cef1f8cbc407cd540c176.tar.gz libvpx-5a3e3c6d3fa308066e2cef1f8cbc407cd540c176.tar.bz2 libvpx-5a3e3c6d3fa308066e2cef1f8cbc407cd540c176.zip | |
Adaptive txfm size selection depending on residual sse/variance
This commit enables an adaptive transform size selection method
for speed -6. It uses largest transform size when the sse is more
than 4 times of variance, i.e., most energy is compacted in the
DC coefficient. Otherwise, use the default TX_8X8. It improves
the compression efficiency for rtc set of speed -6 by 0.8%, no
speed change observed.
Change-Id: Ie6ed1e728ff7bf88ebe940a60811361cdd19969c
Diffstat (limited to 'vp9/encoder/vp9_encodeframe.c')
| -rw-r--r-- | vp9/encoder/vp9_encodeframe.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/vp9/encoder/vp9_encodeframe.c b/vp9/encoder/vp9_encodeframe.c index d1c882443..5387e681c 100644 --- a/vp9/encoder/vp9_encodeframe.c +++ b/vp9/encoder/vp9_encodeframe.c @@ -2390,7 +2390,7 @@ static TX_MODE select_tx_mode(const VP9_COMP *cpi) { rd_opt->tx_select_threshes[frame_type][TX_MODE_SELECT] ? ALLOW_32X32 : TX_MODE_SELECT; } else if (cpi->sf.tx_size_search_method == USE_TX_8X8) { - return ALLOW_8X8; + return TX_MODE_SELECT; } else { unsigned int total = 0; int i; |