VisionQuery v0

Minimal monorepo prototype:

• Backend: FastAPI + YOLO-World v2 (frame-sampled detection only)
• Frontend: React (Vite)

Supports:

• Upload video
• Enter comma-separated prompt (e.g. person, knife)
• Sample frames at requested FPS and run YOLO-World v2 detection
• Show detections and click-to-seek in the video player

No tracking, embeddings, async jobs, or exports.

Run (Backend)

cd visionquery/backend
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
uvicorn main:app --reload --port 8000

The first run will download the YOLO-World weights automatically.

Run (Frontend)

cd visionquery/frontend
npm install
npm run dev

Open the app at http://localhost:5173.

Optional: set backend URL

By default the frontend calls http://localhost:8000.

You can override:

export VITE_BACKEND_URL="http://localhost:8000"

VisionQuery

ChatGPT for Video Surveillance