[FFmpeg-user] Why are half the frames "duplicate" ?

[sean darcy]

On 9/30/18 10:45 AM, sean darcy wrote:
> On 9/29/18 5:15 PM, Carl Eugen Hoyos wrote:
>> 2018-09-29 22:46 GMT+02:00, sean darcy <seandarcy2 at gmail.com>:
>>> On 9/28/18 7:29 PM, Carl Eugen Hoyos wrote:
>>>> 2018-09-28 20:58 GMT+02:00, sean darcy <seandarcy2 at gmail.com>:
>>>>
>>>>> Also, I used ffprobe:
>>>>>
>>>>> ffprobe -show_frames -select_streams v -i in.mpg  2>&1 | grep -c
>>>>> interlaced_frame=1
>>>>> 2052
>>>>> ffprobe -show_frames -select_streams v -i in.mpg  2>&1 | grep -c
>>>>> interlaced_frame=0
>>>>> 30347
>>>>>
>>>>> This means there are at least some interlaced frames
>>>>
>>>> No, unfortunately ffprobe cannot tell you (the information above
>>>> is mostly unrelated to the question if the content is interlaced,
>>>> telecined or progressive).
>>>>
>>>> Either use the idet filter or do a visual inspection.
>>>>
>>>> Carl Eugen
>>>> _______________________________________________
>>>
>>> Never used idet filter before, so don't know how to make use of the 
>>> info.
>>>
>>> ffmpeg -vf idet -frames:v 5000 -an -f rawvideo -y /dev/null -i  in.mpg
>>> ffmpeg version git-snapshot-20180928-RPMFusion Copyright (c) 2000-2018
>>> the FFmpeg developers
>>>     built with gcc 8 (GCC)
>>>     configuration: --prefix=/usr --bindir=/usr/bin
>>> --datadir=/usr/share/ffmpeg --incdir=/usr/include/ffmpeg
>>> --libdir=/usr/lib64 --mandir=/usr/share/man --arch=x86_64
>>> --optflags='-O2 -ffast-math -march=native -ftree-vectorize
>>> -fomit-frame-pointer -pipe -fPIC' --enable-bzlib --disable-crystalhd
>>> --enable-fontconfig --enable-frei0r --enable-gcrypt --enable-gnutls
>>> --enable-ladspa --enable-libass --enable-libbluray --enable-libcdio
>>> --enable-libfdk-aac --enable-nonfree --enable-indev=jack
>>> --enable-libfreetype --enable-libfribidi --enable-libgsm
>>> --enable-libmp3lame --enable-openal --enable-opencl --disable-libopencv
>>> --enable-opengl --enable-libopenjpeg --enable-libopus --enable-libpulse
>>> --enable-libsoxr --enable-libspeex --enable-libtheora --enable-libvorbis
>>> --enable-libv4l2 --enable-libvpx --enable-libwebp --enable-libx264
>>> --enable-libx265 --enable-avfilter --enable-postproc --enable-pthreads
>>> --disable-static --enable-shared --enable-gpl --enable-version3
>>> --enable-nonfree --disable-debug --enable-stripping
>>> --shlibdir=/usr/lib64 --enable-runtime-cpudetect
>>>     libavutil      56. 19.101 / 56. 19.101
>>>     libavcodec     58. 31.101 / 58. 31.101
>>>     libavformat    58. 18.103 / 58. 18.103
>>>     libavdevice    58.  4.104 / 58.  4.104
>>>     libavfilter     7. 33.100 /  7. 33.100
>>>     libswscale      5.  2.100 /  5.  2.100
>>>     libswresample   3.  2.100 /  3.  2.100
>>>     libpostproc    55.  2.100 / 55.  2.100
>>> [mpeg2video @ 0x1eef200] Invalid frame dimensions 0x0.
>>>       Last message repeated 10 times
>>> Input #0, mpeg, from 'in.mpg':
>>>     Duration: 26:09:34.50, start: 1305.384367, bitrate: 91 kb/s
>>>       Stream #0:0[0x1e0]: Video: mpeg2video (Main), yuv420p(tv,
>>> progressive), 720x480 [SAR 8:9 DAR 4:3], Closed Captions, 29.97 fps,
>>> 59.94 tbr, 90k tbn, 59.94 tbc
>>>       Stream #0:1[0x81]: Audio: ac3, 48000 Hz, 5.1(side), fltp, 448 kb/s
>>>       Stream #0:2[0x80]: Audio: ac3, 48000 Hz, 5.1(side), fltp, 448 kb/s
>>>       Stream #0:3[0x1bf]: Data: dvd_nav_packet
>>>       Stream #0:4[0x22]: Subtitle: dvd_subtitle
>>>       Stream #0:5[0x21]: Subtitle: dvd_subtitle
>>>       Stream #0:6[0x20]: Subtitle: dvd_subtitle
>>> Stream mapping:
>>>     Stream #0:0 -> #0:0 (mpeg2video (native) -> rawvideo (native))
>>> Press [q] to stop, [?] for help
>>> Output #0, rawvideo, to '/dev/null':
>>>     Metadata:
>>>       encoder         : Lavf58.18.103
>>>       Stream #0:0: Video: rawvideo (I420 / 0x30323449), yuv420p, 720x480
>>> [SAR 8:9 DAR 4:3], q=2-31, 124291 kb/s, 29.97 fps, 29.97 tbn, 29.97 tbc
>>>       Metadata:
>>>         encoder         : Lavc58.31.101 rawvideo
>>> More than 1000 frames duplicated= 2191360kB time=00:02:24.44
>>> bitrate=124280.6kbits/s dup=867 drop=0 speed=72.2x
>>> frame= 5000 fps=2162 q=-0.0 Lsize= 2531250kB time=00:02:46.83
>>> bitrate=124291.7kbits/s dup=1001 drop=0 speed=72.1x
>>> video:2531250kB audio:0kB subtitle:0kB other streams:0kB global
>>> headers:0kB muxing overhead: 0.000000%
>>> [Parsed_idet_0 @ 0x21f7dc0] Repeated Fields: Neither:  4000 Top:     0
>>> Bottom:     0
>>> [Parsed_idet_0 @ 0x21f7dc0] Single frame detection: TFF:     0 BFF:
>>> 0 Progressive:  3894 Undetermined:   106
>>> [Parsed_idet_0 @ 0x21f7dc0] Multi frame detection: TFF:     0 BFF:     0
>>> Progressive:  4000 Undetermined:     0
>>>
>>> As I read this, the idet filter only saw 4000 out of the 5000 frames, of
>>> which 106 couldn't be classified. What about the 1001 duplicated 
>>> frames ?
>>
>> This indicates soft-telecine of progressive original.
>>
>> Please test with output option "-r 24000/1001". If that
>> leads to (many) duplicated and dropped frames, there
>> are ways to work-around iirc.
>>
>> Carl Eugen
> 
> You are the man !
> 
> Ran the entire clip at 24000/1001, only 11 duplicated frames.
> What did you see in the idet output that told you that the progressive 
> original (~24fps, correct ? ) had been telecined ?
> 
I've been googling soft telecine. Some have suggested using "-re" as the 
frame rate. Why would 24000/1001 be preferred ?