GTC 2020: How to Build a Multi-Camera Media Server for AI Processing on Jetson
After clicking “Watch Now” you will be prompted to login or join.
Click “Watch Now” to login or join the NVIDIA Developer Program.
How to Build a Multi-Camera Media Server for AI Processing on Jetson
Carlos Rodriguez, RidgeRun | Jennifer Caballero, RidgeRun
We'll build a simple multi-camera media server for AI processing on a Jetson Board and demonstrate how, by using GStreamer Daemon, Interpipes, and DeepStream, you can develop a scalable and robust prototype to capture from multiple cameras using GMSL2 Virtual Channels. Besides achieving real-time deep learning inference, the server is completely dynamic. We'll show its flexibility by triggering actions such as taking a snapshot, recording a video, or starting a network streaming when a specific prediction is made. By the end of this session you'll have acquired the framework basics that will allow you to scale to your specific multimedia and AI needs.