Rakshit Agrawal, Director of Research and Development at Camio, presents the “How Containerization Unblocks Barriers to Fast, Easy Deployment of AI-Driven Visual Recognition Applications” tutorial at the May 2021 Embedded Vision Summit.
Recent progress in machine learning algorithms has been meteoric. Yet their implementation in real-world embedded applications crawls at a snail’s pace, in large part due to inconsistent and inefficient software architectures.
In this session, Agrawal introduces the concept of software containerization, explains how it works, and examines how a containerized deployment methodology at the edge can circumvent barriers that currently hinder the deployment of new and improved ML algorithms. He also explores the trade-offs and limitations of using containerization, and shows how his company has used containerization to enable new capabilities like mask detection and license plate recognition to be quickly deployed on to existing cameras.
See here for a PDF of the slides.