CSCL: Critical Semantic-Consistent Learning for Unsupervised Domain Adaptation
文献类型：会议论文
| 作者 | Dong JH(董家华)1,2,3 ; Cong Y(丛杨)2,3; Sun G(孙干)2,3; Liu YY(刘宇阳)1,2,3; Xu XW(徐晓伟)4
|
| 出版日期 | 2020 |
| 会议日期 | August 23-28, 2020 |
| 会议地点 | Glasgow, United kingdom |
| 关键词 | Unsupervised domain adaptation Semantic segmentation Adversarial learning Reinforcement learning Pseudo label |
| 页码 | 745-762 |
| 英文摘要 | Unsupervised domain adaptation without consuming annotation process for unlabeled target data attracts appealing interests in semantic segmentation. However, 1) existing methods neglect that not all semantic representations across domains are transferable, which cripples domain-wise transfer with untransferable knowledge; 2) they fail to narrow category-wise distribution shift due to category-agnostic feature alignment. To address above challenges, we develop a new Critical Semantic-Consistent Learning (CSCL) model, which mitigates the discrepancy of both domain-wise and category-wise distributions. Specifically, a critical transfer based adversarial framework is designed to highlight transferable domain-wise knowledge while neglecting untransferable knowledge. Transferability-critic guides transferability-quantizer to maximize positive transfer gain under reinforcement learning manner, although negative transfer of untransferable knowledge occurs. Meanwhile, with the help of confidence-guided pseudo labels generator of target samples, a symmetric soft divergence loss is presented to explore inter-class relationships and facilitate category-wise distribution alignment. Experiments on several datasets demonstrate the superiority of our model. © 2020, Springer Nature Switzerland AG. |
| 产权排序 | 1 |
| 会议录 | Computer Vision – ECCV 2020 - 16th European Conference, 2020, Proceedings
 |
| 会议录出版者 | Springer Science and Business Media Deutschland GmbH |
| 会议录出版地 | Berlin |
| 语种 | 英语 |
| ISSN号 | 0302-9743 |
| ISBN号 | 978-3-030-58597-6 |
| 源URL | [http://ir.sia.cn/handle/173321/28357]  |
| 专题 | 沈阳自动化研究所_机器人学研究室 |
| 通讯作者 | Cong Y(丛杨) |
| 作者单位 | 1.University of Chinese Academy of Sciences, Beijing 100049, China 2.Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110016, China 3.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China 4.Department of Information Science, University of Arkansas at Little Rock, Little Rock, United States |
| 推荐引用方式 GB/T 7714 | Dong JH,Cong Y,Sun G,et al. CSCL: Critical Semantic-Consistent Learning for Unsupervised Domain Adaptation[C]. 见:. Glasgow, United kingdom. August 23-28, 2020. |
入库方式： OAI收割
来源：沈阳自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。