Intel Edge Insights for Vision

Intel® Edge Insights for Vision is a comprehensive software suite designed for deploying computer vision and deep-learning inference workloads at the edge. It offers a set of pre-integrated components, optimized for Intel® architecture, and can be implemented either as a containerized solution or a standalone runtime.

This platform is engineered to support various monitoring applications across industries such as healthcare, worker safety, and retail, utilizing advanced deep learning models, facial recognition, and analytics capabilities.

This document serves as a guide to help users get started with Edge Insights for Vision on Linux systems. It provides detailed installation instructions, an overview of its architecture and components, and practical tutorials for implementing various computer vision tasks.

Key Features and Tutorials

• Installation: Comprehensive step-by-step instructions for installing Edge Insights for Vision on Linux, including system requirements and setup procedures.
• How it Works: An explanation of the software's architecture, detailing its modules, the Intel® Distribution of OpenVINO™ toolkit, and container orchestration technologies like Docker* and K3s*.
• Tutorials: Practical, hands-on guides covering object detection (single and multi-object) on various hardware configurations (CPU, GPU, Intel® Vision Accelerator) and a multi-camera social distancing detection solution.
• Troubleshooting: Guidance for resolving common installation issues and considerations for specific network environments (e.g., PRC).
• Release Notes: Information on the latest updates, new features, and known issues.

For further assistance and advanced tools, consult the Intel® Edge Software Configurator Documentation.

For community support and issue tracking, visit the Intel Community Support Forum.

Details on specific OpenVINO™ toolkit issues can be found at GitHub.