Transformer-based Generative Models

Vision-Transformer (ViT)

Adaptation of the Transformer architecture for image processing, replacing convolutions with attention mechanisms on image patches.

← Geri