Distributed Facial Recon
summary
A distributed face‑recognition system built with Go services on Kubernetes using gRPC+protobuf contracts, NSQ queues and a Python dlib-based recognizer; images land in MySQL-backed storage (DB generates IDs, transactions prevent collisions), queued via NSQ stateful set, processed by scalable workers, and checked by a circuit breaker. The face service is an ~800MB container, persistent volumes are noted but not wired in the demo, and the pipeline successfully recognizes known people in the live demo.
AI-written summary · about
chapters (12)
- 0:00Start
- 2:07Introduction
- 3:39Technologies
- 19:19Face recognition
- 21:02Microservice Architecture
- 22:31Data flow
- 30:24Distributed System
- 33:04Use cases
- 33:46Dog belly rubbing appointment
- 34:17Mood tracker
- 34:34Super Simple Architecture
- 34:58Real purpose
related talks
57mFoundations of Facial Recognition
39mNodeJS, ML, K8s and Unethical Face Recognition
21mBuilding your own dystopic surveillance state with Python
31mFace Recognition
21mLeveraging Python for Real-Time Image processing
46mImage Recognition with Deep Learning From prototype to production
27mDevelop Computer Vision Apps With Go
11mSaFace: Towards Scenario-aware Face Recognition via Edge Computing System