fansunqi/AKeyS

Agentic Keyframe Search for Video Question Answering

22
/ 100
Experimental

This project helps video analysts, researchers, and content managers quickly find answers to specific questions about video content. It takes a video and a question as input, then identifies only the most relevant frames to provide an accurate answer, reducing the need to manually review long videos. The primary users are those who need to extract concise information from extensive video datasets.

No commits in the last 6 months.

Use this if you need to efficiently answer questions about the content of long videos without watching the entire footage.

Not ideal if your main goal is general video summarization or transcription rather than answering specific queries about events or objects within the video.

video-analysis content-moderation research-assist media-intelligence video-indexing
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

16

Forks

Language

Python

License

MIT

Last pushed

Apr 07, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/fansunqi/AKeyS"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.