Multiscale Vision Transformers

Multiscale Vision Transformers (MViT) for video and image recognition, by connecting the seminal idea of multiscale feature hierarchies with transformer models.

Data and Resources

Cite this as

Haoqi Fan, Bo Xiong, Karttikeya Mangalam, Yanghao Li, Zhicheng Yan, Jitendra Malik, Christoph Feichtenhofer (2024). Dataset: Multiscale Vision Transformers. https://doi.org/10.57702/opc3c3ie

DOI retrieved: December 2, 2024

Additional Info

Field Value
Created December 2, 2024
Last update December 2, 2024
Defined In https://doi.org/10.48550/arXiv.2104.11227
Author Haoqi Fan
More Authors
Bo Xiong
Karttikeya Mangalam
Yanghao Li
Zhicheng Yan
Jitendra Malik
Christoph Feichtenhofer
Homepage https://github.com/facebookresearch/SlowFast