[FFmpeg-cvslog] hevc_mvs: avoid deriving tmvp in amvp

Christophe Gisquet git at videolan.org
Wed Jul 23 21:42:12 CEST 2014


ffmpeg | branch: master | Christophe Gisquet <christophe.gisquet at gmail.com> | Wed Jul 23 18:27:44 2014 +0200| [373677f95816924a3790c174e83ec2e378962c44] | committer: Michael Niedermayer

hevc_mvs: avoid deriving tmvp in amvp

Reduces on a sequence number of calls from 933685 to 586271.

Reviewed-by: Mickaël Raulet <mraulet at insa-rennes.fr>
Signed-off-by: Michael Niedermayer <michaelni at gmx.at>

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=373677f95816924a3790c174e83ec2e378962c44
---

 libavcodec/hevc_mvs.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/libavcodec/hevc_mvs.c b/libavcodec/hevc_mvs.c
index 52cfb29..1d4c002 100644
--- a/libavcodec/hevc_mvs.c
+++ b/libavcodec/hevc_mvs.c
@@ -751,7 +751,8 @@ scalef:
         mvpcand_list[numMVPCandLX++] = mxB;
 
     //temporal motion vector prediction candidate
-    if (numMVPCandLX < 2 && s->sh.slice_temporal_mvp_enabled_flag) {
+    if (numMVPCandLX < 2 && s->sh.slice_temporal_mvp_enabled_flag &&
+        mvp_lx_flag == numMVPCandLX) {
         Mv mv_col;
         int available_col = temporal_luma_motion_vector(s, x0, y0, nPbW,
                                                         nPbH, ref_idx,



More information about the ffmpeg-cvslog mailing list