TPC-ViT: Token Propagation Controller for Efficient Vision Transformers

Vision transformers (ViTs) have achieved promising results on a variety of Computer Vision tasks, however their quadratic complexity in the number of input tokens has limited their application specially in resource-constrained settings. The proposed TPC-ViT is evaluated based on the ImageNet-1K image classification dataset.

Data and Resources

Cite this as

Wentao Zhu (2024). Dataset: TPC-ViT: Token Propagation Controller for Efficient Vision Transformers. https://doi.org/10.57702/5vptpj53

DOI retrieved: December 16, 2024

Additional Info

Field Value
Created December 16, 2024
Last update December 16, 2024
Author Wentao Zhu
Homepage https://arxiv.org/abs/2205.11155