vp10: make segmentation probs use generic probability model.
Locate them (code-wise) in frame_context, and have them be updated as any other probability using the subexp forward and adaptive bw updates. See issue 1040 point 1. TODOs: - real-world default probabilities - why is counts sometimes NULL in the decoder? Does that mean bw adaptivity updates only work on some frames? (I haven't looked very closely yet, maybe this is a red herring.) Change-Id: I23b57b4e5e7574b75f16eb64823b29c22fbab42e
Showing
- vp10/common/entropymode.c 26 additions, 0 deletionsvp10/common/entropymode.c
- vp10/common/entropymode.h 13 additions, 0 deletionsvp10/common/entropymode.h
- vp10/common/onyxc_int.h 3 additions, 0 deletionsvp10/common/onyxc_int.h
- vp10/common/pred_common.h 3 additions, 3 deletionsvp10/common/pred_common.h
- vp10/common/seg_common.h 5 additions, 3 deletionsvp10/common/seg_common.h
- vp10/common/thread_common.c 11 additions, 0 deletionsvp10/common/thread_common.c
- vp10/decoder/decodeframe.c 23 additions, 6 deletionsvp10/decoder/decodeframe.c
- vp10/decoder/decodemv.c 47 additions, 10 deletionsvp10/decoder/decodemv.c
- vp10/encoder/bitstream.c 50 additions, 7 deletionsvp10/encoder/bitstream.c
- vp10/encoder/encoder.c 8 additions, 4 deletionsvp10/encoder/encoder.c
- vp10/encoder/encoder.h 2 additions, 0 deletionsvp10/encoder/encoder.h
- vp10/encoder/segmentation.c 77 additions, 28 deletionsvp10/encoder/segmentation.c
- vp10/encoder/segmentation.h 1 addition, 1 deletionvp10/encoder/segmentation.h
Loading
Please register or sign in to comment