FFmpeg Time Lapse from Sources with Long Frozen Tail End

Question

My source for inputs into FFmpeg is either one AVI file or a concat of many of them. Either way my resulting timelapse file has a long tail of frames that are a repeat of a single frame. It's like a very long freeze frame is resulting on the end of my output file.

I'm on an older Mac Pro so I can't update my FFmpeg. I have a laptop that has a newer version and I get the same there as well. I'm not sure if it's because my source files are AVI or not.

ffmpeg version 4.1.1 Copyright (c) 2000-2019 the FFmpeg developers
    built with Apple LLVM version 8.0.0 (clang-800.0.42.1)
    configuration: --prefix=/usr/local/Cellar/ffmpeg/4.1.1 --enable-shared --enable-pthreads --enable-version3 --enable-hardcoded-tables --enable-avresample --cc=clang --host-cflags=-I/System/Library/Frameworks/JavaVM.framework/Versions/Current/Headers/ --host-ldflags= --enable-ffplay --enable-gnutls --enable-gpl --enable-libaom --enable-libbluray --enable-libmp3lame --enable-libopus --enable-librubberband --enable-libsnappy --enable-libtesseract --enable-libtheora --enable-libvorbis --enable-libvpx --enable-libx264 --enable-libx265 --enable-libxvid --enable-lzma --enable-libfontconfig --enable-libfreetype --enable-frei0r --enable-libass --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-librtmp --enable-libspeex --enable-videotoolbox --disable-libjack --disable-indev=jack --enable-libaom --enable-libsoxr
    libavutil      56. 22.100 / 56. 22.100
    libavcodec     58. 35.100 / 58. 35.100
    libavformat    58. 20.100 / 58. 20.100
    libavdevice    58.  5.100 / 58.  5.100
    libavfilter     7. 40.101 /  7. 40.101
    libavresample   4.  0.  0 /  4.  0.  0
    libswscale      5.  3.100 /  5.  3.100
    libswresample   3.  3.100 /  3.  3.100
    libpostproc    55.  3.100 / 55.  3.100
[avi @ 0x7fa82080c800] sample size (1) != block align (2)
Guessed Channel Layout for Input Stream #0.1 : mono
Input #0, avi, from 'concat:16460001.AVI|16460002.AVI|16460003.AVI|16460004.AVI|16460005.AVI|16460006.AVI|16460007.AVI|16460008.AVI|16460009.AVI|16460010.AVI|16460011.AVI|16460012.AVI|16460013.AVI|16460014.AVI|16460015.AVI|16460016.AVI|16460017.AVI|16460018.AVI|16460019.AVI|16460020.AVI|16460021.AVI':
    Duration: 00:10:02.00, start: 0.000000, bitrate: 365923 kb/s
        Stream #0:0: Video: mjpeg (MJPG / 0x47504A4D), yuvj420p(pc, bt470bg/unknown/unknown), 1280x720, 30 fps, 30 tbr, 30 tbn, 30 tbc
        Stream #0:1: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 32000 Hz, mono, s16, 512 kb/s
Stream mapping:
    Stream #0:0 -> #0:0 (mjpeg (native) -> h264 (libx264))
    Stream #0:1 -> #0:1 (pcm_s16le (native) -> aac (native))
Press [q] to stop, [?] for help
[libx264 @ 0x7fa82082cc00] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.1 Cache64
[libx264 @ 0x7fa82082cc00] profile High, level 3.1
[libx264 @ 0x7fa82082cc00] 264 - core 155 r2917 0a84d98 - H.264/MPEG-4 AVC codec - Copyleft 2003-2018 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2 threads=12 lookahead_threads=2 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=16 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00
Output #0, mp4, to 'out.mp4':
    Metadata:
        encoder         : Lavf58.20.100
        Stream #0:0: Video: h264 (libx264) (avc1 / 0x31637661), yuvj420p(pc, progressive), 1280x720, q=-1--1, 16 fps, 16384 tbn, 16 tbc
        Metadata:
            encoder         : Lavc58.35.100 libx264
        Side data:
            cpb: bitrate max/min/avg: 0/0/0 buffer size: 0 vbv_delay: -1
        Stream #0:1: Audio: aac (LC) (mp4a / 0x6134706D), 32000 Hz, mono, fltp, 69 kb/s
        Metadata:
            encoder         : Lavc58.35.100 aac
frame= 1962 fps=1.2 q=-1.0 Lsize=  136725kB time=03:24:07.00 bitrate=  91.5kbits/s dup=0 drop=365448 speed=7.41x
video:31548kB audio:103624kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 1.149437%
[libx264 @ 0x7fa82082cc00] frame I:10    Avg QP:18.74  size: 62176
[libx264 @ 0x7fa82082cc00] frame P:514   Avg QP:21.24  size: 30528
[libx264 @ 0x7fa82082cc00] frame B:1438  Avg QP:22.74  size: 11121
[libx264 @ 0x7fa82082cc00] consecutive B-frames:  1.4%  2.0%  2.0% 94.6%
[libx264 @ 0x7fa82082cc00] mb I  I16..4:  2.1% 97.2%  0.7%
[libx264 @ 0x7fa82082cc00] mb P  I16..4:  2.0% 33.3%  0.1%  P16..4: 37.7% 12.0% 10.4%  0.0%  0.0%    skip: 4.5%
[libx264 @ 0x7fa82082cc00] mb B  I16..4:  0.7% 11.5%  0.0%  B16..8: 29.1%  4.8%  1.3%  direct:12.0%  skip:40.6%  L0:47.5% L1:42.9% BI: 9.6%
[libx264 @ 0x7fa82082cc00] 8x8 transform intra:94.3% inter:83.0%
[libx264 @ 0x7fa82082cc00] coded y,uvDC,uvAC intra: 73.4% 66.7% 8.5% inter: 25.9% 46.8% 2.2%
[libx264 @ 0x7fa82082cc00] i16 v,h,dc,p: 17% 33% 26% 24%
[libx264 @ 0x7fa82082cc00] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 13% 23% 51%  1%  2%  1%  4%  1%  2%
[libx264 @ 0x7fa82082cc00] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 17% 21% 19%  5%  8%  7% 13%  5%  5%
[libx264 @ 0x7fa82082cc00] i8c dc,h,v,p: 46% 29% 23%  2%
[libx264 @ 0x7fa82082cc00] Weighted P-Frames: Y:47.5% UV:19.3%
[libx264 @ 0x7fa82082cc00] ref P L0: 45.0% 13.8% 20.6% 15.7%  5.0%
[libx264 @ 0x7fa82082cc00] ref B L0: 61.1% 30.4%  8.5%
[libx264 @ 0x7fa82082cc00] ref B L1: 85.8% 14.2%
[libx264 @ 0x7fa82082cc00] kb/s:2107.54
[aac @ 0x7fa82081ea00] Qavg: 130.084

I've read this thread and tried to follow it:

Create time lapse video from other video

Here are some trials I've attempted before, all with the same output:

# ffmpeg -y -i $mov -vf framestep=10,setpts=N/FRAME_RATE/TB,fps=2 -r 30 $out
# ffmpeg -y -i $mov -vf framestep=10,setpts=.01*PTS -r 30 $out
# ffmpeg -y -i $mov -vf framestep=10,setpts=.1*PTS -r 30 $out
# ffmpeg -y -i "concat:16460001.AVI|16460002.AVI|16460003.AVI|16460004.AVI|16460005.AVI|16460006.AVI|16460007.AVI|16460008.AVI|16460009.AVI|16460010.AVI|16460011.AVI|16460012.AVI|16460013.AVI|16460014.AVI|16460015.AVI|16460016.AVI|16460017.AVI|16460018.AVI|16460019.AVI|16460020.AVI|16460021.AVI" -vf framestep=10,setpts=.05*PTS -r 30 $out
ffmpeg -y -i "concat:16460001.AVI|16460002.AVI|16460003.AVI|16460004.AVI|16460005.AVI|16460006.AVI|16460007.AVI|16460008.AVI|16460009.AVI|16460010.AVI|16460011.AVI|16460012.AVI|16460013.AVI|16460014.AVI|16460015.AVI|16460016.AVI|16460017.AVI|16460018.AVI|16460019.AVI|16460020.AVI|16460021.AVI" -r 16 -filter:v "setpts=0.01*PTS" out.mp4

Am I overlooking something? I'm trying to speed up the inputs into a single file to quickly review incoming security footage. How can I do this without the super long useless tail at the end?

I want to avoid audio altogether. Did some trials for two source files and those frames are still there, deaf and otherwise. `ffmpeg -y -an -i "concat:16460001.AVI|16460002.AVI" -r 16 -filter:v "setpts=0.01*PTS" $out_deaf ffmpeg -y -i "concat:16460001.AVI|16460002.AVI" -r 16 -filter:v "setpts=0.01*PTS" $out` — Rich_F, Jul 04 '19 at 18:23
Even single file, deaf and otherwise, still retaining 10:02 in length, with the tail of it being a repeated frame. — Rich_F, Jul 04 '19 at 18:25
One hour or so: https://www.dropbox.com/s/f9dywnnn2wflxoq/16460021.AVI?dl=0 — Rich_F, Jul 04 '19 at 19:25
Just realized the deaf version maintains the audio track for some reason. See above. — Rich_F, Jul 04 '19 at 20:07
Unless you're using a recent version of ffmpeg, `-an` should be after the inputs. — Gyan, Jul 04 '19 at 20:13
Got a test to work...`-an` was placed in the wrong location. — Rich_F, Jul 04 '19 at 20:13
Now `-an` isn't working: `ffmpeg -y -i /Volumes/RUN_SD16/DCIM/10090704/14080016.AVI -vf framestep=15,setpts=N/30/TB,fps=30 -an /users/rich/desktop/final_run.mp4` — Rich_F, Jul 04 '19 at 21:38

score 1 · Accepted Answer · answered Jul 05 '19 at 05:30

1

Use

ffmpeg -i 16460021.AVI -vf framestep=15,setpts=N/30/TB -r 30 -an out.mp4

I get an output 1/15th the length of the original. When using framestep=X, setpts should be of the form N/X/TB if you wish to avoid frame duplication. Don't use fps when compressing video duration as it uses old stream duration from before compression (a bug). Skip audio, since that will result in original length.

answered Jul 05 '19 at 05:30

Gyan

85,394
9
169
201

OK that worked for a single file. I'm working with an old version, and I wanted to see the faster route of the two approaches. Taking out the audio fixed the bad tail. Thank you. – Rich_F Jul 05 '19 at 05:37

FFmpeg Time Lapse from Sources with Long Frozen Tail End

1 Answers1