Watch on youtube.com


As you can see, “pixscope” overlays analysis results onto input video, but there is no way to control the font size of this (and also font color), so if you use this for videos like 1920x1080 which is common, you should not be able to read it at all because the letters are too small (and blue font is too inconspicuous).

If you intend to use this seriously for video analysis, it is better to reduce the size of the input video even at the expense of accuracy, as we do here. Otherwise there is no practicality at all.

#! /bin/sh
ifnb="`basename \"${ifn}\" .mp4`"
pref="`basename $0 .sh`"

"/c/Program Files/ffmpeg-4.1-win64-shared/bin/ffmpeg" -y \
    -i "${ifn}" -filter_complex "
" -map '[v]' -an "${pref}_${ifnb}.mp4"