System overview

The flow is simple:

Camera (RTSP) ──→ Tensor Dynamics Watch ──→ AI model ──→ Live log
                                       (count / anomaly / threat)

• You provide: an RTSP URL per camera.
• The app does: pulls frames once per second and sends them to the AI.
• The AI returns: person count, scene description, and whether to alert (the tile turns red).

Capacity

Up to 8 cameras at once (4 × 2 grid). You don't need to know AI or write code. You just do two things: tell it where the cameras are, and pick an AI engine.