Towards Rich Feature Discovery with Class Activation Maps Augmentation for Person Re-Identification | |
Yang, Wenjie1,2,4,5; Huang, Houjing1,2,4,5; Zhang, Zhang1,2,4,5; Chen, Xiaotang1,2,4,5; Huang, Kaiqi1,2,3,4,5 | |
2019-06 | |
会议日期 | June 16-20 |
会议地点 | Long Beach, United States |
英文摘要 | The fundamental challenge of small inter-person variation requires Person Re-Identification (Re-ID) models to capture sufficient fine-grained features. This paper proposes to discover diverse discriminative visual cues without extra assistance, e.g., pose estimation, human parsing. Specifically, a Class Activation Maps (CAM) augmentation model is proposed to expand the activation scope of baseline Re-ID model to explore rich visual cues, where the backbone network is extended by a series of ordered branches which share the same input but output complementary CAM. A novel Overlapped Activation Penalty is proposed to force the current branch to pay more attention to the image regions less activated by the previous ones, such that spatial diverse visual features can be discovered. The proposed model achieves state-of-the-art results on three Re-ID datasets. Moreover, a visualization approach termed ranking activation map (RAM) is proposed to explicitly interpret the ranking results in the test stage, which gives qualitative validations of the proposed method |
语种 | 英语 |
内容类型 | 会议论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/44900] |
专题 | 智能系统与工程 |
通讯作者 | Huang, Kaiqi |
作者单位 | 1.Center for Research on Intelligent Perception and Computing 2.National Laboratory of Pattern Recognition 3.CAS Center for Excellence in Brain Science and Intelligence Technology 4.University of Chinese Academy of Sciences 5.Institute of Automation, Chinese Academy of Sciences |
推荐引用方式 GB/T 7714 | Yang, Wenjie,Huang, Houjing,Zhang, Zhang,et al. Towards Rich Feature Discovery with Class Activation Maps Augmentation for Person Re-Identification[C]. 见:. Long Beach, United States. June 16-20. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
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。
修改评论