How to cut and merge mp4 video segments without re-encoding?

Question

I have some screencast video files in mp4 container (h.264/mp3 codecs) that I want at times merge (i.e. C.mp4 = A.mp4+B.mp4) or at times cut removing some intervals (e.g. D.mp4 = A.mp4[0,320{seconds}]+A.mp4[325,340]+C.mp4).

I am now using kdenlive, it woks but it re-encode everything and it seems a bit too much for such a task.

Is there a simpler way to do that without re-encoding (even, and perhaps preferred, command line or stable Python/Julia/R/.. library way) ?

I am thinking at something like PDFtk for the pdfs :-))

PS: Following suggestions from a related question I did try ffmpeg -ss 00:00:05 -to 00:00:10 -i test1.mp4 test2.mp4 but I got the error:

Option to (record or transcode stop time) cannot be applied to input url test1.mp4 -- you are trying to apply an input option to an output file or vice versa. Move this option before the file it belongs to.

I would like to be able to specify several start/stop times from a single or perhaps multiple input files.

EDIT: More I look, more I find complicate solutions. My files are exactly the same codec/container/resolution..

Answer 1

Well, this doesn't solve the problem of the re-encoding, but at least it is an handy interface. Just use it (in Linux) with vcat -i inputfile1,inputfile2[start-end],... -o <outputfile> (it needs python3 and the moviepy package)

#!/usr/bin/python3

import sys, getopt, re

def printerror(errormsg):
  print("*** vcat - concatenate video segments using moviepy ***\n")
  print("ERROR:", errormsg,"\n")
  print("Usage: vcat -i inputfile1,inputfile2[start-end],... -o <outputfile>")
  print("Optional start and end points should be given in seconds. If files have spaces should be quoted (e.g. \"input file.mp4[5-30]\").")

try:
 from moviepy.editor import VideoFileClip, concatenate_videoclips
except ImportError:
 printerror("You don't seem to have moviepy installed. Install it with `pip install moviepy`.")
 exit(1)

def main(argv):
    inputfiles_arg = ''
    outputfile = ''
    try:
        opts, args = getopt.getopt(argv,"hi:o:",["input=","output="])
    except getopt.GetoptError as err:
        printerror(str(err))
        sys.exit(2)
    for opt, arg in opts:
        if opt == '-h':
            printerror("")
            sys.exit()
        elif opt in ("-i", "--input"):
            inputfiles_arg = arg
        elif opt in ("-o", "--output"):
            outputfile = arg
    if outputfile =='':
        printerror("Output file not specified")

    iFiles = inputfiles_arg.split(',')

    clips=[]
    for iFile in iFiles:
        subclip = re.search(r"\[([0-9\-\.]+)\]", iFile)
        if subclip is None:
            clips.append(VideoFileClip(iFile))
        else:
            dims = subclip.group(1).split('-')
            if len(dims) != 2:
                printerror("If a specific segment of a file is specified, this should be given as [startseconds-endseconds]")
            iFile = iFile.replace("["+subclip.group(1)+"]",'')
            clips.append(VideoFileClip(iFile).subclip(float(dims[0]),float(dims[1])))

    f = concatenate_videoclips(clips)
    f.write_videofile(outputfile)


if __name__ == "__main__":
   main(sys.argv[1:])

Disclaimer: I don't use Python by a while, so sure there could be better approaches.. bdw it's really nice to be able to do what you want to do in a few hours because of the enormous documentation you can find on the net about python..

Answer 2

IT will help merge movie1..3.mp4

For cut can add -ss 00:00:XX -to 00:00:XX for -i file

ffmpeg -y -hide_banner  \
 -i "movie1.mp4" -c copy -bsf:v h264_mp4toannexb -an -f mpegts \
 intermediate1.ts

ffmpeg -y -hide_banner  \
 -i "movie2.mp4" -c copy -bsf:v h264_mp4toannexb -an -f mpegts \
 intermediate2.ts

ffmpeg -y -hide_banner  \
 -i "movie3.mp4" -c copy -bsf:v h264_mp4toannexb -an -f mpegts \
 intermediate3.ts

ffmpeg -y  -hide_banner \
 -i "concat:intermediate1.ts|intermediate2.ts|intermediate3.ts" \
 -i wind.wav \
 -c:v copy -c:a aac -b:a 64k \
 "mergedmovie.mp4"