Learn about Mask2Former, a universal image segmentation method, in a 12-minute video that breaks down the three main subtasks of image segmentation - instance, semantic, and panoptic segmentation. Follow along with a practical image example demonstrating how Mask2Former handles all three segmentation types simultaneously. Explore vision transformer technology, implementation details, and relevant datasets while preparing for hands-on PyTorch GUI development covered in subsequent content.
Overview
Syllabus
Introduction
Vision Technology
Mask2Former
Datasets
Taught by
Discover AI