MIA-Former: Efficient and Robust Vision Transformers via Multi-Grained Input-Adaptation.

Zhongzhi Yu Yonggan Fu Sicheng Li Chaojian Li Yingyan Lin

Published in: AAAI (2022)

Keyphrases

computationally efficient
fine grained
vision system
real time
real world
computer vision
face recognition
highly efficient
databases
machine learning
search engine
input data
cost effective
visual perception