Metron project tackles use case of measurement of objects. It is a great platform to test various computer vision approaches or as well to be used for commercial Proof of Concepts.
Switch into Metron dev branch to see the latest commits.
💥 efficient input video source streaming engine 💥 various input video sources - video files, camera device 💥 fully configurable using Facebook Research - Hydra framework 💥 run on a bare metal or using Docker Compose 💥 deployable on Linux, MacOS, Windows
1. Project Installation
2. Architecture
3. Development Notes
During roadmap timeline gaps, the work is conducted on Metron AI research track. Take a look on Peer Repositories links section.
- setup Metron Shine component SW skeleton
- add message broker component (Redis)
- testing of the message broker component
- document message broker component
- setup Metron Shine configuration
- implement Video Receiver module
- implement Results Renderer module
- implement Results Streamer module
- testing of Metron Shine
- develop Metron Shine demo app
- dockerize Metron Shine
- documentation of Metron Shine
- software architecture
- setup Metron Conduit software skeleton
- add video file source option to Metron Conduit
- add usb cam source option to Metron Conduit
- complete basic Video Streamer functionality of Metron Conduit
- dockerize Metron Conduit
- setup Docker Compose deployment skeleton
- add Hydra configuration framework
-
camera stream input module -
camera stream calibration