https://github.com/jes/gpt4o-video

Analysing videos with gpt4o
https://github.com/jes/gpt4o-video

Last synced: 12 months ago
JSON representation

Analysing videos with gpt4o

• Host: GitHub
• URL: https://github.com/jes/gpt4o-video
• Owner: jes
• Created: 2024-09-06T20:40:01.000Z (almost 2 years ago)
• Default Branch: master
• Last Pushed: 2024-09-07T15:12:48.000Z (almost 2 years ago)
• Last Synced: 2025-03-23T17:49:23.543Z (over 1 year ago)
• Language: Perl
• Homepage: https://incoherency.co.uk/videogrep/
• Size: 24.4 KB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
## GPT4o video processing

We split the video up into frames at 1fps, stick them in a GPT4o Vision context, and then have the AI answer questions about the video.

## Example

On this video:

https://www.youtube.com/watch?v=ahvqPh8dJ8s

[![asciicast](https://asciinema.org/a/ca45P50bJAs12wbVEcG9xHguy.svg)](https://asciinema.org/a/ca45P50bJAs12wbVEcG9xHguy)

### How to Use

1. **Interacting with the Go program**

   - Run the Go program to interact with the split frames.

     ```sh

     go run . -video foo.mp4 -fps 1

     ```

2. **Non-interactive mode**

   - Run the Go program to interact with the split frames.

     ```sh

     go run . -video foo.mp4 -fps 1 -prompt-file prompt.txt

     ```