mbzuai-oryx/Video-LLaVA

PG-Video-LLaVA: Pixel Grounding in Large Multimodal Video Models

30
/ 100
Emerging

This tool helps you understand what's happening in videos by letting you ask questions in plain language and pointing to specific objects. You provide a video, and it answers your questions, highlighting or 'grounding' the objects you mention. This is useful for anyone who needs to quickly extract detailed information from video content, such as researchers analyzing human behavior or media analysts reviewing news footage.

262 stars. No commits in the last 6 months.

Use this if you need to precisely locate and understand objects or events in a video based on conversational prompts, especially when audio context is important.

Not ideal if your primary need is simple video transcription or if you require real-time, ultra-low-latency object detection for live streams.

video-analysis content-moderation media-monitoring behavioral-research surveillance-review
No License Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 8 / 25
Community 10 / 25

How are scores calculated?

Stars

262

Forks

12

Language

Python

License

Last pushed

Aug 05, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/mbzuai-oryx/Video-LLaVA"

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