Opened 11 years ago
Closed 9 years ago
#3147 closed defect (fixed)
H264 - Wrong field order
Reported by: | Underground78 | Owned by: | |
---|---|---|---|
Priority: | normal | Component: | avcodec |
Version: | git-master | Keywords: | h264 |
Cc: | Blocked By: | ||
Blocking: | Reproduced by developer: | no | |
Analyzed by developer: | no |
Description
Summary of the bug:
This sample is top field first but ffmpeg seems to detect it as bottom field first.
How to reproduce:
% ffprobe.exe -show_frames VIDEO_Panasonic_AG_HPX_600.MXF ffprobe version N-58134-g571a99d Copyright (c) 2007-2013 the FFmpeg developers built on Nov 16 2013 18:01:42 with gcc 4.8.2 (GCC) configuration: --enable-gpl --enable-version3 --disable-w32threads --enable-avisynth --enable-bzlib --enable-fontconfig --enable-frei0r --enable-gnutls --enable-iconv --enable-libass --enable-libbluray --enable-libcaca --enable-libfreetype --enable-libgsm --enable-libilbc --enable-libmodplug --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libopus --enable-librtmp --enable-libschroedinger --enable-libsoxr --enable-libspeex --enable-libtheora --enable-libtwolame --enable-libvidstab --enable-libvo-aacenc --enable-libvo-amrwbenc --enable-libvorbis --enable-libvpx --enable-libwavpack --enable-libx264 --enable-libxavs --enable-libxvid --enable-zlib libavutil 52. 53.100 / 52. 53.100 libavcodec 55. 43.100 / 55. 43.100 libavformat 55. 21.100 / 55. 21.100 libavdevice 55. 5.100 / 55. 5.100 libavfilter 3. 91.100 / 3. 91.100 libswscale 2. 5.101 / 2. 5.101 libswresample 0. 17.104 / 0. 17.104 libpostproc 52. 3.100 / 52. 3.100 [mxf @ 047cc640] material track 2: no corresponding source track found [h264 @ 047de940] Overread VUI by 5 bits Last message repeated 1 times [mxf @ 047cc640] Stream #0: not enough frames to estimate rate; consider increasing probesize Input #0, mxf, from 'VIDEO_Panasonic_AG_HPX_600.MXF': Metadata: uid : 7ddd2bbf-701b-4733-b70f-1ce186dfc582 generation_uid : 2badcdf8-04f8-48df-a76c-52f7ea9a15fa company_name : Panasonic product_name : P2 product_version : 2.0 product_uid : 060e2b34-0401-0109-0e0e-010101000001 modification_date: 2012-12-12 15:03:24 timecode : 01:04:54:18 Duration: 00:00:41.72, start: 0.000000, bitrate: 113670 kb/s Stream #0:0: Video: h264 (High 4:2:2 Intra), yuv422p10le(pc, bt709), 1920x1080 [SAR 1:1 DAR 16:9], 25 fps, 25 tbr, 25 tbn, 50 tbc [h264 @ 048422e0] Overread VUI by 5 bits [FRAME] media_type=video key_frame=1 pkt_pts=N/A pkt_pts_time=N/A pkt_dts=N/A pkt_dts_time=N/A pkt_duration=1 pkt_duration_time=0.040000 pkt_pos=32768 pkt_size=568832 width=1920 height=1080 pix_fmt=yuv422p10le sample_aspect_ratio=1:1 pict_type=I coded_picture_number=0 display_picture_number=0 interlaced_frame=1 top_field_first=0 repeat_pict=0 [/FRAME] ...
Original ticket: https://trac.mpc-hc.org/ticket/3732.
Change History (4)
comment:1 by , 11 years ago
Component: | undetermined → avcodec |
---|---|
Keywords: | h264 added |
comment:3 by , 10 years ago
I also stumbled over this problem a few weeks ago. From looking at the source and some small debugging it seems that MediaInfo derives the field order from the picture "bottom_field_flag":
Recommendation H.264 (https://www.itu.int/rec/dologin_pub.asp?lang=e&id=T-REC-H.264-200305-S!!PDF-E&type=items) page 80:
"""
bottom_field_flag equal to 1 specifies that the slice is part of a coded
bottom field. bottom_field_flag equal to 0 specifies that the picture is a
coded top field. When this syntax element is not present for the current
slice, it shall be inferred to be equal to 0.
"""
FFmpeg relies only on the information in h->sei_pic_struct
. When trying to use h->picture_structure
I stumble over the issue that FFmpeg defaults to h->picture_structure = PICT_FRAME
when the bottom_field_flag is absent instead of h->picture_structure = PICT_BOTTOM_FIELD
as mentioned in the specs. But then I am unsure how to differ between interlaced and non-interlaced data.
comment:4 by , 9 years ago
Resolution: | → fixed |
---|---|
Status: | new → closed |
Fixed in 2b6567722a48656c526ca2d6c1dcab88be3c18f7 by Thomas Mundt
Is there an application that correctly shows top-field-first?
h->sps.pic_struct_present_flag
is set,h->sei_pic_struct
is set toSEI_PIC_STRUCT_FRAME
, mxf does not setdescriptor->field_dominance
afaict.