FFmpeg Aufbaukurs, 13. März 2025
Cours avancé de FFmpeg, 14 mars 2025

AV Preservation by reto.ch und Ødegård & Bernard Restaurierung Klg
Reto Kromer, Kamilla Ødegård, Eléonore Bernard


COMMANDS

CLI BASICS
----------

# set the working space
#
# on Linux/Mac/Windows Terminal or WSL/Windows locally
cd Desktop

# on Windows on OneDrive Cloud
cd OneDrive

If not working in the specific folder, the whole path of the file has to be inserted in the command! This is useful, if you do not want to save the output file in the same folder as the input file.


FFMPEG HELP
-----------

ffmpeg -h

ffmpeg -codecs

ffmpeg -decoders

ffmpeg -h decoder=aac

ffmpeg -encoders

ffmpeg -h encoder=libx264

ffmpeg -filters

ffmpeg -pix_fmts


CASE STUDY - CUT COMMANDS
-------------------------

# Cut command without reencoding
ffmpeg -i input_file -ss hh:mm:ss -to hh:mm:ss -c:v copy -c:a copy output_file

# Cut command with reencoding of MKV/ffv1
ffmpeg -accurate_seek -i input_file.mkv -ss hh:mm:ss -to hh:mm:ss -c:v ffv1 -level 3 -coder 1 -context 1 -g 1 -slices 24 -slicecrc 1 -c:a pcm_s16le output_file.mkv

# Cut command with reencoding of MP4/H.264
ffmpeg -accurate_seek -i input_file.mp4 -ss hh:mm:ss -to hh:mm:ss -c:v libx264 -preset veryslow -crf 18 -pix_fmt yuv420p -c:a aac output_file.mp4


QUALITY CONTROL
---------------

# Generate MD5 checksums
# On Mac:
md5 input_file
# On Windows:
certutil -hashfile input_file MD5

# generate checksums at frame level
ffmpeg -i input_file -f framemd5 output_file_framemd5.txt

# compare checksum files
#
# on Linux/Mac/Windows Terminal or WSL
diff -s mandelbrot_avi_framemd5.txt mandelbrot_mov_framemd5.txt
# on Windows
fc mandelbrot_avi_framemd5.txt mandelbrot_mov_framemd5.txt

--> can also be used for other Exports and Textfiles

# FFprobe
ffprobe [input_options] input_file
ffprobe -show_format -show_streams -print_format json input_file.avi > input_file.txt

# MediaInfo
mediainfo file.ext
mediainfo —Details=1 file.ext
mediainfo —Output=JSON file.ext

# Splitscreen
ffmpeg -i input_file_1 -i input_file_2 -filter_complex "[0] crop=iw/2:ih:0:0 [left]; [1] crop=iw/2:ih:iw/2:0 [right]; [left][right] hstack [out]" -map "[out]" output_file

# Differenzsignal
ffmpeg -i input_file_1 -i input_file_2 -filter_complex "[1] format=yuva444p, lut=c3=128, negate [1_with_alpha]; [0][1_with_alpha] overlay [out]“ -map „[out]" output_file


DVD files: merge and transcode to MP4
-------------------------------------

# Transcode several VOB files to one MPEG file (not deinterlaced file):
ffmpeg -i "concat:VTS_01_1.VOB|VTS_01_2.VOB|VTS_01_3.VOB|VTS_01_4.VOB|VTS_01_5.VOB" -c:v copy -c:a copy concat.VOB

# Transcode several VOB files to one MP4:
ffmpeg -i "concat:VTS_01_1.VOB|VTS_01_2.VOB|VTS_01_3.VOB" -b:v 1500k -r 30 -vcodec h264 -strict -2 -acodec aac -ar 44100 -f mp4 output_file.mp4

ffmpeg -i "concat:input_file_1|input_file_2|input_file_3" -c:v libx264 -c:a aac output_file.mp4

# all Audio und Video channels berücksichtigen:
ffmpeg -i "concat:input_file_1|input_file_2" -map 0:v -map 0:a -c:v libx264 -c:a aac output_file.mp4 
# Source: https://amiaopensource.github.io/ffmprovisr/#dvd_to_file

# includes deinterlacing SD:
ffmpeg -i concat:VTS_02_1.VOB\|VTS_03_1.VOB\|VTS_04_1.VOB -filter:v "idet, bwdif, scale=1440:1080:flags=lanczos" -c:v libx264 -c:a aac output_H264.mp4

# includes deinterlacing und SD -> HD:
ffmpeg -i concat:VTS_01_1.VOB\|VTS_01_2.VOB\|VTS_01_3.VOB -filter:v "idet, bwdif, scale=1440:1080:flags=lanczos, pad=1920:1080:240:0“ -c:v libx264 -c:a aac output_H264.mp4


DPX to MKV
----------

# for single images
ffmpeg -f image2 -framerate 24 -i input_file_%08d.dpx -c:v ffy1 -level 3 -threads 8 -coder 1 -context 1 -g 1 -slices 24 -slicerc 1 -start_number 00086400 -c:a copy output_file.mkv

# for image stream / video
ffmpeg -i input_file.mxf -c:v ffv1 -level 3 -threads 8 -coder 1 -context 1 -g 1 -slices 24 -slicecrc 1 -c:a copy output_file.mkv

# for single images to MP4
ffmpeg -f image2 -framerate 24 -i input_file_%06d.extension -c:v libx264 -preset veryslow -crf 18 -pix_fmt yuv420p output_file


ProRes to viewing copy
----------------------

ffmpeg -i inputfile.mov -pix_fmt yuv420p -c:v libx264 -preset veryslow -crf 30 -movflags +faststart+write_colr outputfile_ProRes_H264.mp4
# source: https://avpres.net/FFmpeg/im_H264.html 


Uncompressed to viewing copy
----------------------------

ffmpeg -i input_file.MOV -c:v libx264 -preset veryslow -crf 18 -pix_fmt yuv420p -c:a aac output_file.MP4

# deinterlacing
ffmpeg -i input_file -c:v libx264 -pix_fmt yuv420p -filter:v "idet, bwdif" output_file.MP4
# OR
ffmpeg -i input_file -c:v libx264 -preset veryslow -crf 18 -pix_fmt yuv420p -c:a aac -filter:v "idet, bwdif" output_file.MP4

# 4:3 in 16:9 mit Pillarbox
ffmpeg -i input_file -c:v libx264 -filter:v "yadif, scale=1440:1080:flags=lanczos, pad=1920:1080:(ow-iw)/2:(oh-ih)/2, format=yuv420p" output_file


Changing Scan mode
------------------

# deinterlacing
ffmpeg -i input_file -c:v libx264 -pix_fmt yuv420p -filter:v "idet, bwdif" output_file

# deinterlacing (with another type of argument)
ffmpeg -i input_file -c:v libx264 -vf "yadif,format=yuv420p" output_file


Changing Color space
--------------------

# change colormatrix:
ffmpeg -i input_file -c:v libx264 -vf colormatrix=src:dst output_file

# example change Rec. 601 (used for SD) to Rec. 709 (used for HD):
ffmpeg -i input_file -c:v libx264 -vf colormatrix=bt601:bt709 output_file

# Color space for PAL:
ffmpeg -i input_file -c:v libx264 -color_primaries bt470bg -color_trc bt709 -colorspace bt470bg output_file

# Color space for NTSC:
ffmpeg -i input_file -c:v libx264 -color_primaries smpte170m -color_trc bt709 -colorspace smpte170m output_file


Changing image resolution
-------------------------

# SD ➔ HD with Pillarbox (https://avpres.net/FFmpeg/SD_HD.html)
ffmpeg -i input_file -filter:v "pad=ih*16/9:ih:(ow-iw)/2:(oh-ih)/2" -c:a copy output_file

ffmpeg -i input_file -filter:v "scale=1440:1080:flags=lanczos, pad=1920:1080:240:0" -c:a copy output_file

# Display Aspect Ratio ändern 
ffmpeg -i input_file -c:a copy -c:v copy -aspect 4:3 output_file


Adding to video: Watermark, Timecode, Logo, Credits
---------------------------------------------------

# https://amiaopensource.github.io/ffmprovisr/#text_watermark
# show which fonts are installed
#
# on macOS
ls /Library/Fonts

# on Windows
dir \Windows\Fonts

# add watermark
ffmpeg -i Dufay_master_H264.mp4 -filter:v "drawtext=text='watermark':fontfile='/Library/Fonts/Arial.ttf':fontsize=35:fontcolor=white:alpha=0.25:x=(w-text_w)/2:y=(h-text_h)/2" with_watermark.mp4

# add watermark (source: https://amiaopensource.github.io/ffmprovisr/#text_watermark)
ffmpeg -i input_file -vf drawtext="fontfile=font_path:fontsize=font_size:text=watermark_text:fontcolor=font_color:alpha=0.4:x=(w-text_w)/2:y=(h-text_h)/2" output_file

# add watermark example
ffmpeg -i input_file -filter:v "drawtext=text='watermark':fontfile='/Library/Fonts/Arial.ttf':fontsize=35:fontcolor=white:alpha=0.25:x=(w-text_w)/2:y=(h-text_h)/2" output_file

# add logo
ffmpeg -i input_file.mp4 -i logo.png -filter_complex "overlay=10:main_h-overlay_h-10" with_logo.mp4

# add timecode
ffmpeg -i input_file -filter:v drawtext="timecode=starting_timecode:rate=timecode_rate:fontfile=font_path:fontsize=font_size:fontcolor=font_colour:box=1:boxcolor=box_colour:x=(w-text_w)/2:y=h/1.2" output_file

# add timecode example
ffmpeg -i input_file.mp4 -filter:v drawtext="timecode='01\:00\:00\:00':rate=25:fontfile='/Library/Fonts/Arial.ttf':fontsize=35:fontcolor=white:x=(w-text_w)/2:y=h/1.2" with_timecode.mp4

# add credits
# idea: concat command for two videos, if credits are present as a video file

# add subtitles
ffmpeg -i output_archive.mkv -filter:v "subtitles=subtitles.srt" -c:v libx264 -preset veryslow -crf 18 -pix_fmt yuv420p -c:a aac -ar 48k -movflags +faststart+write_colr output_streaming.mp4