Getting Started
As a concrete example to become familiar with PySceneDetect, let's use the following short clip from the James Bond movie, GoldenEye (Copyright © 1995 MGM):
https://www.youtube.com/watch?v=OMgIPnCnlbQ
You can download the clip from here (may have to right-click and save-as, put the video in your working directory as goldeneye.mp4).
Content-Aware Detection
In this case, we want to split this clip up into each individual scene - at each location where a fast cut occurs. This means we need to use content-aware detecton mode (detect-content) or adaptive mode (detect-adaptive). If the video instead contains fade-in/fade-out transitions you want to find, you can use detect-threshold instead.
Using the following command, let's run PySceneDetect on the video, and also save a scene list CSV file and some images of each scene:
scenedetect --input goldeneye.mp4 detect-content list-scenes save-images
Running the above command, in the working directory, you should see a file goldeneye.scenes.csv, as well as individual frames for the start/middle/end of each scene as goldeneye-XXXX-00/01.jpg. The results should appear as follows:
| Scene # | Start Time | Preview |
|---|---|---|
| 1 | 00:00:00.000 |  |
| 2 | 00:00:03.754 |  |
| 3 | 00:00:08.759 |  |
| 4 | 00:00:10.802 |  |
| 5 | 00:00:15.599 |  |
| 6 | 00:00:27.110 |  |
| 7 | 00:00:34.117 |  |
| 8 | 00:00:36.536 |  |
| ... | ... | ... |
| 18 | 00:01:06.316 |  |
| 19 | 00:01:10.779 |  |
| 20 | 00:01:18.036 |  |
| 21 | 00:01:19.913 |  |
| 22 | 00:01:21.999 |  |
Splitting Video into Clips
The split-video command can be used to automatically split the input video using ffmpeg or mkvmerge. For example:
scenedetect -i goldeneye.mp4 detect-content split-video
Type scenedetect help split-video for a full list of options which can be specified for video splitting, including high quality mode (-hq/--high-quality) or copy mode (-c/--copy).
Finding Optimal Threshold/Sensitivity
If the default threshold of 27 does not produce correct results, we can determine the proper threshold by generating a statistics file with the -s / --stats option:
scenedetect --input goldeneye.mp4 --stats goldeneye.stats.csv detect-content
We can then plot the values of the content_val column:
The peaks in values correspond to the scene breaks in the input video. In some cases the threshold may need to be raised or lowered accordingly.