Workflow for face-detection social network

Video-process -> S3 -> lambda -> Google ML -> S3 -> lambda(Social-Graph) -> Dynamo

Cloud infrastructure list: AWS S3 AWS lambda AWS dynamo Google ML engine

Todos:

1. Front-end API design (X)
2. Distributed ML Engine
3. End-to-end test
4. Demo page - d3 social networking graph
5. Load Test

Notes:

1. Video-processing happens in front-end sensor
2. No Final back-end requirement.