Duration
14 hours (usually 2 days including breaks)
Requirements
- An understanding of deep neural networks
- Python and C programming experience
Audience
- Developers
- Data scientists
Overview
Video analytics refers to the technology and techniques used to process a video stream. A common application would be capturing and identifying live video events through motion detection, facial recognition, crowd and vehicle counting, etc.
This instructor-led, live training (online or onsite) is aimed at developers who wish to build hardware-accelerated object detection and tracking models to analyze streaming video data.
By the end of this training, participants will be able to:
- Install and configure the necessary development environment, software and libraries to begin developing.
- Build, train, and deploy deep learning models to analyze live video feeds.
- Identify, track, segment and predict different objects within video frames.
- Optimize object detection and tracking models.
- Deploy an intelligent video analytics (IVA) application.
Format of the Course
- Interactive lecture and discussion.
- Lots of exercises and practice.
- Hands-on implementation in a live-lab environment.
Course Customization Options
- To request a customized training for this course, please contact us to arrange.
Course Outline
Introduction
Understanding Hardware Accelerated Decoding Methods
Overview of NVidia DeepStream SDK
Setting up the Development Environment
Preparing a Video Feed
Processing a Video Feed
Training a Deep Learning Model
How Transfer Learning Works
Improving the Model’s Accuracy Through Transfer Learning
Developing a Neural Network Model to Track Moving Objects
Running a Video Analytics Inference Engine
Deploying the Inference Engine
Integrating a Deep Learning Model with an Application
Deploying an Intelligent Video Analytics (IVA) Application
Monitoring the Application
Optimizing the Inference Engine and Application
Troubleshooting
Summary and Conclusion