It provides real-time object detection, semantic video search, and event indexing designed for on-premise and air-gapped deployments.

Core capabilities include:

• Real-time object detection (YOLO-based) • Semantic search over recorded footage (CLIP-based embeddings) • Multi-camera ingestion and indexing • Configurable detection thresholds • Designed for local deployment (no cloud dependency)

We’re releasing this as part of building an open ecosystem around reliable, real-world video intelligence systems.

Feedback, contributions, and technical discussion are welcome.