Uploaded January 2024 | Updated September 2026, 1 week ago

---

Colab notebook: colab.research.google.com/drive/1S5rDmP1dSxjZd2VoJDLLjne-Bu9Vw6Xp?usp=sharing

Join this channel to get access to perks:
youtube.com/channel/UCaZuPdmZ380SFUMKHVsv_AA/join

Learn more about machine learning for image makers by signing up at https://mailchi.mp/da905fbd76ee/machine-learning-artists

Join the Artificial Images Slack channel: join.slack.com/t/ml-images/shared_invite/zt-1erimzv04-pLZPnFktsJvBMVy8n7Z4tw

artificial-images.com
patreon.com/bustbright
bustbright.com
instagram.com/dvsmethid
twitter.com/dvsch
dvschultz.github.io/design

Chapters:
00:00 PySceneDetect notebook overview
00:45 Connect Colab and disable GPU
01:30 Mount Google Drive
03:00 Install PySceneDetect and YouTube downloader
04:30 Find a YouTube video ID
06:00 Download and organize the trailer
07:30 Configure shot detection and output clips
09:00 Run scene detection on Groundhog Day
11:15 Review generated scene clips
12:00 Process an uploaded video instead
15:45 Detect scenes in NYC footage
17:15 Sample the Groundhog Day cuts
18:45 Memento reshuffle example
20:15 Diagnose false cuts and flicker