- Acquire knowledge on Deep Neural Networks (DNN) techniques within the deep learning workflow to solve a real-world video analytics problems
- Learn how to use Tensor RT to optimize developed DNN
- Learn about general optimization technique of IVA algorithm on GPUs
- Learn about INT8 for DNN optimization knowledge and techniques
- Learn about GPU multi-thread safe application & management techniques
Please bring your laptop to participate in hands-on exercises. A GPU in your laptop is not required.
No previous GPU programming experience is required for the hands-on. However, beginner-level C and Linux experience will be expected.
To familiarize yourself with the course, we would recommend to watch a few short (approx. 5 min) YouTube videos on introductory GPU programming topics. These CUDACasts can be found here. Also, please take time to register at our CUDA developer for more updates and information.
- The session fee is RM100 until further notice (normal rate RM220).
- Participant can make the payment via the online registration page.
- Cancellation of registration and full refund are only allowed within 48 hours prior to the program. Any cancellations received less than 48 hours will not be refunded.
Intelligent Video Analytics (IVA) Using Deep Learning
When
Where
About the Event
Ready to transform everyday video into valuable insight and decision-making? NVIDIA Intelligent Video Analytics (IVA) makes it easy for companies to create robust security and surveillance solutions - from cameras to appliances to servers – using the power of deep learning. The session is especially useful for companies to boost sales, increase customer satisfaction, lower theft, and boost public safety. This session will cover the advanced topics of Deep Learning, including IVA application development methods, optimization techniques, Deep Neural Networks (DNN) training and deployment environment recommendations. The course is suitable for developers, engineers and data scientists who would like to learn more and get trained on optimizing and deploying neural networks to solve real-world problems via video analytics, security and surveillance disciplines.
Key Takeaways
Itinerary
8.30am: Registration
9.00am: Deep Learning Techniques & Tensor RT 2.0, and Enhancement
10.00am: Break
10.15am: Optimization Technique and INT8 Technique
11.45am: Break
12.00pm: CPU part of Memory Pooling & Camera stream optimization technique
1.00pm: Break
2.00pm: GPU Multi-thread safe application & Management Technique
3.30pm: Break
3.45pm: Profiling & Debugging
4.45pm: End