Stand-Alone Self-Attention in Vision Models

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

NeurIPS, pp. 68-80, 2019.

Cited by: 12|Bibtex|Views91
EI
Other Links: academic.microsoft.com|dblp.uni-trier.de
Keywords:

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