#19 ViT: An Image is Worth 16x16 Words

#deeplearning #machinelearning #ml #research #ai #artificialintelligence #yannickilcher #ai

The paper is about the first time when transformers were successfully introduced in computer vision.

OUTLINE
0:00 - Channel updates!
1:11 - Introduction
5:11 - ViT Architecture
18:10 - Results
31:25 - Yannic's views
36:00 - GitHub implementation
38:00 - Final Remarks

USEFUL LINKS

GitHub noob implementation: https://github.com/AdityaKane2001/noo...
Paper: https://arxiv.org/abs/2010.11929
Code: https://github.com/google-research/vi...
Annotated paper: https://drive.google.com/file/d/1IZcW...
Yannic's video:    • An Image is Worth 16x16 Words: Transf...  
The AI Epiphany video:    • Vision Transformer (ViT) - An image i...  

ABSTRACT
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.

AUTHORS
Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, Neil Houlsby

SAHIL KHOSE
YouTube:    / sahilkhose  
Website: https://sahilkhose.github.io/
Twitter:   / sahilkhose  
LinkedIn:   / sahilkhose  
GitHub: https://github.com/sahilkhose
Email: [email protected]