Stand-Alone Self-Attention in Vision Models

Prajit Ramachandran
Prajit Ramachandran
Niki Parmar
Niki Parmar
Irwan Bello
Irwan Bello
Anselm Levskaya
Anselm Levskaya

ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 32 (NIPS 2019), 2019.

Cited by: 26|Bibtex|Views110
EI
Other Links: dblp.uni-trier.de|arxiv.org

Abstract:

Convolutions are a fundamental building block of modern computer vision systems. Recent approaches have argued for going beyond convolutions in order to capture long-range dependencies. These efforts focus on augmenting convolutional models with content-based interactions, such as self-attention and non-local means, to achieve gains on a ...More

Code:

Data:

Full Text
Your rating :
0

 

Tags
Comments