Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2021
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2110.14240 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915174177308672 |
|---|---|
| author | Liao, Haojin Song, Xiaolin Zhao, Sicheng Zhang, Shanghang Yue, Xiangyu Yao, Xingxu Zhang, Yueming Xing, Tengfei Xu, Pengfei Wang, Qiang |
| author_facet | Liao, Haojin Song, Xiaolin Zhao, Sicheng Zhang, Shanghang Yue, Xiangyu Yao, Xingxu Zhang, Yueming Xing, Tengfei Xu, Pengfei Wang, Qiang |
| contents | The Visual Domain Adaptation (VisDA) 2021 Challenge calls for unsupervised domain adaptation (UDA) methods that can deal with both input distribution shift and label set variance between the source and target domains. In this report, we introduce a universal domain adaptation (UniDA) method by aggregating several popular feature extraction and domain adaptation schemes. First, we utilize VOLO, a Transformer-based architecture with state-of-the-art performance in several visual tasks, as the backbone to extract effective feature representations. Second, we modify the open-set classifier of OVANet to recognize the unknown class with competitive accuracy and robustness. As shown in the leaderboard, our proposed UniDA method ranks the 3rd place with 48.49% ACC and 70.8% AUROC in the VisDA 2021 Challenge. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2110_14240 |
| institution | arXiv |
| publishDate | 2021 |
| record_format | arxiv |
| spellingShingle | 3rd Place Solution for VisDA 2021 Challenge -- Universally Domain Adaptive Image Recognition Liao, Haojin Song, Xiaolin Zhao, Sicheng Zhang, Shanghang Yue, Xiangyu Yao, Xingxu Zhang, Yueming Xing, Tengfei Xu, Pengfei Wang, Qiang Computer Vision and Pattern Recognition The Visual Domain Adaptation (VisDA) 2021 Challenge calls for unsupervised domain adaptation (UDA) methods that can deal with both input distribution shift and label set variance between the source and target domains. In this report, we introduce a universal domain adaptation (UniDA) method by aggregating several popular feature extraction and domain adaptation schemes. First, we utilize VOLO, a Transformer-based architecture with state-of-the-art performance in several visual tasks, as the backbone to extract effective feature representations. Second, we modify the open-set classifier of OVANet to recognize the unknown class with competitive accuracy and robustness. As shown in the leaderboard, our proposed UniDA method ranks the 3rd place with 48.49% ACC and 70.8% AUROC in the VisDA 2021 Challenge. |
| title | 3rd Place Solution for VisDA 2021 Challenge -- Universally Domain Adaptive Image Recognition |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2110.14240 |