Advantages of nested transformers for computer vision
Advantages of nested transformers for computer vision
Aggregating Nested Transformers
arXiv paper abstract https://arxiv.org/abs/2105.12723
arXiv PDF paper https://arxiv.org/pdf/2105.12723.pdf
... explore ... nesting basic local transformers on non-overlapping image blocks and aggregating them in a hierarchical manner.
... leads ... to ... a simplified architecture with minor code changes upon the original vision transformer and obtains improved performance compared to existing methods.
... converges faster and requires much less training data
... Training a NesT with 6M parameters from scratch on CIFAR10 achieves 96% accuracy using a single GPU, setting a new state of the art for vision transformers.
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comentários