Patterns recognized by deep learning models from visual information have reached unprecedented precision levels in diverse fields. However, most of the state-of-the-art computer vision systems require high computing resources, which in turn makes them need high bandwidth for image or video transferring to central servers. Recently, the significant improvements in GPU, TPU, and CPU components on diverse embedded devices have opened opportunities for new computer vision applications with chances to work in real-time. Nevertheless, inference near where the data is collected still requires strategies to cope with constrained resources to function without delays.  Moreover, efficient systems at the edge must be designed considering power and memory monitoring and estimation. In this tutorial, new opportunities, tools, strategies, and challenges for the deployment of computer vision systems will be explored. Moreover, concrete experiences using OpenVino, TensoRT, ONNX, NVIDIA Jetson Nano, OAK-D and NCS2 will be explained in detail to address the mentioned challenges.

Target Audience

AI researchers and developers, students and academics working on deep learning applications.

Pre-requisites

Knowledge in Python, specifically about virtual environments and libraries.
ML development concepts and TensorFlow
Good understanding of the console/terminal of the operating system they will be working on, whether it's Windows, MacOS, or a Linux distribution.

Technical requirements

Operating System: Windows 10+, Ubuntu 18.04 - 20.04, MacOS Big Sur. Correct functioning of the provided material will not reliable for other OS, but it will not be a limiting factor to attend the tutorial
Git
Visual Studio Code
Anaconda (Participants should know how to create virtual environments and install libraries)