AI Object Detection Integration

I need to implement a feature that detects objects in images and videos using the Gemini model, displaying results in real-time on an overlay.

Key Requirements:
- Real-time detection for videos
- Overlay for displaying detection results

Ideal Skills:
- Proficient in OpenAI model integration
- Experience with Google Cloud Platform
- FastAPI for backend, Next.js for frontend
- UI with Next.js, Tailwind CSS, Shadcn

If you complete this feature in two days, there's potential for ongoing work with my team.