NOTE: My website is currently undergoing construction (migrating to a Github-based solution), and thus the below links do not work (they will once the migration is completed).
In the meantime, you can view a cached version of the pages below using the Internet Archive's Wayback Machine.
This repository contains the example programs, corresponding to the three-part tutorial on detecting scene changes in videos with Python and OpenCV, which can be found at:
- Simple Threshold/Fade-to-Black Detection
- Optimized/Adaptive Threshold Detection
- Suppressing/Discarding False Scene Changes (coming soon!)
You can get the latest sample code/video from the releases page. All files in the examples/ directory are released into the public domain. In total, there are three programs (part1-threshold.py, part2-adaptive.py, and still under development, part3-content-aware.py) corresponding to the respective part in the tutorial linked above. Usage details for each program can be found in the comment header for that file.
Related to this tutorial is PySceneDetect, a robust OpenCV-based scene detection program.
Copyright (C) 2013-2020 Brandon Castellano. All rights reserved.