NVIDIA Deep Learning Institute: AI for computer vision and digital content creation (Day 2)

The NVIDIA Deep Learning Institute (DLI) will walk you through the fundamentals of deep learning – training neural networks and using results to improve performance and capabilities. Once you’ve learned the basics, you’ll apply deep learning to digital content creation and game development to start creating digital asset with deep learning. This two-day training is designed for developers, data scientists, and researchers – no prior deep learning experience required.

Outline

Day 1: Fundamentals of Deep Learning for Computer Vision

• Implementing common deep learning workflows, such as image classification and object detection
• Experimenting with data, training parameters, network structure, and other strategies to increase performance and capability
• Deploying your neural networks to start solving real-world problems

Day 2: Deep Learning for Digital Content Creation Using GANs and Autoencoders

• Training a Generative Adversarial Network (GAN) to generate images
• Exploring the architectural innovations and training techniques used to make arbitrary video style transfer
• Training your own denoiser for rendered images