How it works

1. Get video length (in seconds) using ffprobe
2. Calculate frame rate (video_length / desired_image_width)
3. Initialize output_image and temporary_image
   • output_image will be the final barcode image
   • temporary_image will be a 1x1 bitmap containing the average color of the current frame
4. Extract frame from video using ffmpeg and save in temporary_image
5. Draw line on output_image using color from temporary_image
6. Repeat step 4 until we've reached the final frame
7. Save output_image

Usage

positional arguments:
  file                  file to process

optional arguments:
  -h, --help            show this help message and exit
  --output OUTPUT       output file
  --width WIDTH         output image width  (default=1280)
  --height HEIGHT       output image height (default=720)
  --thickness THICKNESS frame thickness     (default=10)

Requirements

• Python3
• Pillow
• FFmpeg
• progress (optional)

Info

• small videos should finish within 1 minute (for default settings)
• large video files are slow (movies can take from 1-5 mintues depending on file size and total length)
• resolution is irrelevant (no need for 4K video files)
• thickness should be a multiple of width but is not enforced
• output_image will be automatically named based on input
  • if --output option is provided with trailing slash then the image will still be automatically named but place in to that directory