A-ViT: Adaptive Tokens for Efficient Vision Transformer.

Hongxu Yin Arash Vahdat Jose M. Alvarez Arun Mallya Jan Kautz Pavlo Molchanov

Published in: CVPR (2022)

Keyphrases

computer vision
cost effective
image processing
vision system
computationally efficient
databases
three dimensional
evolutionary algorithm
adaptive learning