This is the official PyTorch implementation of our paper: Adversarial Bipartite Graph Learning for Video Domain Adaptation
- Python 3.7, PyTorch 1.2, CUDA 10.2
Experiments are conducted on four datasets: UCF-HMDBsmall, UCF-HMDBfull, UCF-Olympic, Kinetics-Gamplay.
The downloaded files need to store in ./dataset.
Pre-extracted features and data lists can be downloaded as,
-
Features
- UCF: download
- HMDB: download
- Olympic: training | validation
-
Data lists
- UCF-Olympic
- UCF: training list | validation list
- Olympic: training list | validation list
- UCF-HMDBsmall
- UCF: training list | validation list
- HMDB: training list | validation list
- UCF-HMDBfull
- UCF: training list | validation list
- HMDB: training list | validation list
- UCF-Olympic
-
Kinetics-Gameplay: please fill this form to request the features and data lists.
The Kinetics-Gameplay dataset is licensed under CC BY-NC-SA 4.0 for non-commercial purposes only.
- training/validation: Run
./script_<DATASET_NAME>_G.shE.g., script_HMDB_Ucf_G.sh
If you find this repository useful, please cite our papers:
@inproceedings{DBLP:conf/mm/LuoHW0B20,
author = {Yadan Luo and
Zi Huang and
Zijian Wang and
Zheng Zhang and
Mahsa Baktashmotlagh},
editor = {Chang Wen Chen and
Rita Cucchiara and
Xian{-}Sheng Hua and
Guo{-}Jun Qi and
Elisa Ricci and
Zhengyou Zhang and
Roger Zimmermann},
title = {Adversarial Bipartite Graph Learning for Video Domain Adaptation},
booktitle = {{MM} '20: The 28th {ACM} International Conference on Multimedia, Virtual
Event / Seattle, WA, USA, October 12-16, 2020},
pages = {19--27},
publisher = {{ACM}},
year = {2020},
url = {https://doi.org/10.1145/3394171.3413897},
doi = {10.1145/3394171.3413897}
}