Opened 11 years ago
Closed 9 years ago
#3077 closed defect (wontfix)
Corruption when decoding experimental hevc stream
Reported by: | Carl Eugen Hoyos | Owned by: | |
---|---|---|---|
Priority: | minor | Component: | avcodec |
Version: | git-master | Keywords: | hevc |
Cc: | mickael raulet | Blocked By: | |
Blocking: | Reproduced by developer: | no | |
Analyzed by developer: | no |
Description
(Sample from xhevc)
No information is shown that a hevc sample cannot be decoded by the current decoder.
$ ffmpeg -i hevc.hm10 -qscale 2 out.avi ffmpeg version N-57366-gfbdc98c Copyright (c) 2000-2013 the FFmpeg developers built on Oct 23 2013 23:51:57 with gcc 4.7 (SUSE Linux) configuration: --enable-gpl libavutil 52. 47.101 / 52. 47.101 libavcodec 55. 38.101 / 55. 38.101 libavformat 55. 19.104 / 55. 19.104 libavdevice 55. 4.100 / 55. 4.100 libavfilter 3. 89.100 / 3. 89.100 libswscale 2. 5.101 / 2. 5.101 libswresample 0. 17.104 / 0. 17.104 libpostproc 52. 3.100 / 52. 3.100 Input #0, hevc, from 'hevc.hm10': Duration: N/A, bitrate: N/A Stream #0:0: Video: hevc, yuv420p, 1280x720, 25 fps, 25 tbr, 1200k tbn, 25 tbc Please use -q:a or -q:v, -qscale is ambiguous Output #0, avi, to 'out.avi': Metadata: ISFT : Lavf55.19.104 Stream #0:0: Video: mpeg4 (FMP4 / 0x34504D46), yuv420p, 1280x720, q=2-31, 200 kb/s, 25 tbn, 25 tbc Stream mapping: Stream #0:0 -> #0:0 (hevc -> mpeg4) Press [q] to stop, [?] for help [hevc @ 0x33a5000] CABAC_MAX_BIN : 100 frame= 933 fps=170 q=2.0 Lsize= 75571kB time=00:00:37.44 bitrate=16535.2kbits/s video:75543kB audio:0kB subtitle:0 global headers:0kB muxing overhead 0.037002%
output looks severely broken.
Attachments (1)
Change History (12)
by , 11 years ago
comment:1 by , 11 years ago
comment:2 by , 11 years ago
http://xhevc.com/en/hevc/decoder/download.jsp
The afd_720p.hm10 sample.
comment:3 by , 11 years ago
Cc: | added |
---|
comment:8 by , 10 years ago
Priority: | normal → minor |
---|---|
Summary: | Corruption when decoding hevc → Corruption when decoding experimental hevc stream |
I suggest closing this bug because: this is not a conforming bitstream, and hm10 is development/experimental version whose bitstreams shall be ignored. Current version is now hm15.
comment:9 by , 10 years ago
I wondered if a better message could be shown (if it is impossible to support these streams): Is "profile 0" the only relevant information the stream contains?
comment:10 by , 10 years ago
profile 0, afaik, doesn't exist and means it could be anything, in particular on how to parse the bitstream. There are a few bits that can be parsed, but nothing that will say "we absolutely can't decode". It's not even a "patch welcome"/"missing feature" here.
"CABAC_MAX_BIN : 100" in the log underlines this: it is not possible and means we've already had bitstream reading errors.
We have something that looks vaguely watchable, and could maybe be fixed, but I can't see how anybody knowing hevc would want to go beyond what mraulet did.
Maybe we can be clearer and say that the decoded output may be total garbage.
comment:11 by , 9 years ago
Resolution: | → wontfix |
---|---|
Status: | new → closed |
could you put the link of where this bitstreams comes from? If it wpp, they fixed it with hm11.