In this module, you'll learn:
  • How to adapt Visual Attention to Image Segmentation problems
  • The 3 main blocks used in Segformer Encoders:
    • Efficient Attention
    • Overlap Patch Embeddings & Mergings
    • Mix Feed Forward Networks & the GELU Activation
  • How to implement and train a Segformer from scratch with PyTorch