Boost your Vision AI Application with Vision Transformer
, Senior Solutions Architect, NVIDIA
, Solutions Architect, NVIDIA
Vision transformers (ViTs) are taking computer vision by storm, offering incredible accuracy and robust solutions for countless industries. However, there are many practical challenges to deploying ViTs, including pre-training, fine tuning, deploying, and managing the complexity of such a large model. Learn how to use Transformer-based vision models for scene understanding using datasets such as Cityscape and Hypersim. First you'll learn how to train a computer vision model, then how to fine-tune the model with Transfer learning, and finally how to deploy the optimized ViT model efficiently in production. You’ll also learn how to leverage ViTs and how to analyze various models. Prerequisite(s):
Python. Fundamentals of Deep Learning. Basic understanding of training frameworks (TF, Pytorch).