various methods to compress a video based on the speed of things moving in it.
reduce each frame to only things that are moving fast enough. condense into the middle of the frame.
python lib/rip.py [-h] -i INPUT [-fd FRAMEDIST] [-min MINMAG] [-max MAXMAG] [-p PADDING] [-o OUTPUT]
normalize the video's speed by making faster things smaller than slower things AND fill in the gaps with nearby color values
python lib/combine_resized_fill.py [-h] -i INPUT [-fd FRAMEDIST] [-ds DOWNSCALE] [-min MINMAG] [-max MAXMAG] [-p PADDING] [-o OUTPUT] [-nf]
normalize the video's speed by making the frame rate of faster things smaller than that of slower things
python lib/combine_sep_speed.py [-h] -i INPUT [-w WINSIZE] [-fd FRAMEDIST] [-d DEPTH] [-o OUTPUT]
only update parts of a video that are moving faster than a given threshold (out of 255, default 50)
python lib/update_only_past_thresh.py [-h] -i INPUT [-w WINSIZE] [-fd FRAMEDIST] [-min MINMAG] [-o OUTPUT]
download and resize a youtube video given a url and maximum dimension
bash tools/download_video.sh "URL" MAX_DIM
- opencv
- tqdm
- numpy
- ffmpeg
- yt-dlp