├── README.md
└── docs
    └── figures
        └── pipeline.png


/README.md:
--------------------------------------------------------------------------------
 1 | # Video Representation Learning with Visual Tempo Consistency
 2 | 
 3 | ![image](./docs/figures/pipeline.png)
 4 | [[Paper](https://arxiv.org/pdf/2006.15489.pdf)]
 5 | [Project Page]
 6 | 
 7 | ## News
 8 | - Full codebae is coming soon
 9 | 
10 | ## Pretained Models
11 | For now, we provide the pretrained models of 3D ResNet-18/50 ([link](https://drive.google.com/drive/folders/1Lgzs6GVkAY-oItwF3_RryGGccxbAZS0h?usp=sharing)). This [repo](https://github.com/decisionforce/TPN) supports the action recognition task.
12 | 
13 | ## Bibtex
14 | ```
15 | @inproceedings{yang2020vthcl,
16 |   title={Video Representation Learning with Visual Tempo Consistency},
17 |   author={Yang, Ceyuan and Xu, Yinghao and Dai, Bo and Zhou, Bolei},
18 |   booktitle={arXiv preprint arXiv:2006.15489},
19 |   year={2020},
20 | }
21 | ```
22 | 


--------------------------------------------------------------------------------
/docs/figures/pipeline.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/decisionforce/VTHCL/368f9adf9dc47a5fd4d236c5aca4d766dca5d883/docs/figures/pipeline.png


--------------------------------------------------------------------------------