Vision Transformer (ViT) attains excellent results compared to state-of-the-art convolutional networks while requiring substantially fewer computational resources to train.
While the Transformer architecture has become the de-facto standard for natural language processing tasks, its applications to computer vision remain limited. In vision, attention is either applied in conjunction with convolutional networks, or used to replace certain components of convolutional networks while keeping their overall structure in place. We show that this reliance on CNNs is not necessary and a pure transformer applied directly to sequences of image patches can perform very well on image classification tasks. When pre-trained on large amounts of data and transferred to multiple mid-sized or small image recognition benchmarks (ImageNet, CIFAR-100, VTAB, etc.), Vision Transformer (ViT) attains excellent results compared to state-of-the-art convolutional networks while requiring substantially fewer computational resources to train.
Alexander Kolesnikov
13 papers
Alexey Dosovitskiy
16 papers
Lucas Beyer
16 papers
M. Minderer
6 papers
Jakob Uszkoreit
7 papers
Dirk Weissenborn
4 papers
Mostafa Dehghani
1 papers
G. Heigold
6 papers
S. Gelly
7 papers