AI-based video analytics can unlock insights across many industries such as smart cities, retail space management, hospital health and safety monitoring, and manufacturing defect detection, among others.

In this course, you will gain the knowledge and skills needed to enable the real-time transformation of raw video data from widely-deployed camera sensors into deep learning-based insights.

Learning Objectives

By participating in this is course, you will:

• Conceptualize the application and anatomy of, as well as challenges facing video AI analytics
• Construct a streaming analytics pipeline with hardware-accelerated components
• Deploy a turnkey video AI application with NVIDIA’s pre-trained inference models
• Apply transfer learning to develop a custom video AI model that is configured for optimal performance
• Measure and improve video AI application performance

Prerequisites:

• Competency in the Python 3 programming language
• Some experience manipulating data using pandas DataFrames.
• Experience with deep networks (specifically variations of CNNs)

Tools, libraries, frameworks used: NVIDIA DeepStream, NVIDIA TAO Toolkit, and NVIDIA TensorRT