Intelligent MultiCamera Identity & Access Security Platform

Inspiration

Traditional security systems detect incidents but struggle to maintain identity across multiple cameras. We wanted to build smarter, proactive physical security.

What it does

It detects people, tracks them across cameras, identifies authorized users, verifies zone access, and alerts security about unknown or unauthorized individuals.

How we built it

We combined YOLOv8, SCRFD, ArcFace, FAISS, ByteTrack, OSNet, and EBIRE with a configurable backend, database, and security dashboard.

Challenges we ran into

Maintaining reliable identity across cameras, handling unknown individuals, balancing recognition accuracy with performance, and designing configurable access policies were our biggest challenges.

Accomplishments that we're proud of

We built a complete multi-camera identity and access pipeline where an unknown person can retain a global identity and be recognized later.

What we learned

We learned that real-world AI systems require more than accurate models. Tracking, identity fusion, system design, configurability, and explainable decisions are equally important.

What's next

We plan to improve Re-ID and recognition accuracy, optimize real-time performance, strengthen configurable policies, and expand SENTINEX into a scalable physical security platform.