Vision Transformers provably learn spatial structure.

Samy Jelassi Michael E. Sander Yuanzhi Li

Published in: CoRR (2022)

Keyphrases

spatial structure
computer vision
image processing
geometric structure
image sequences
vision system