Skip to content

Latest commit

 

History

History
4 lines (4 loc) · 545 Bytes

README.md

File metadata and controls

4 lines (4 loc) · 545 Bytes

JART

Code of my dissertation design, referring to ViT from Dosovitskiy A, Beyer L, Kolesnikov A, et al. JART represents Jiang Action Recognition Transformer, which is a Transformer for image classification tasks. JART has showed excellent performance in UCF101 dataset.

How to run the code

Download UCF-101jpg dataset(not original UCF101) from here, then change train_list_path in train.py to the path where you save your UCF101 dataset. Then just run it!!!