Knowledge Commons of Institute of Automation,CAS
Botom-Up Foreground-Aware Feature Fusion for Person Search | |
Yang, Wenjie1,2,3; Li, Dangwei1,2,3; Chen, Xiaotang1,2,3; Huang, Kaiqi1,2,3,4 | |
2020-10 | |
会议名称 | In Proceedings of the 28th ACM International Conference on Multimedia |
会议日期 | 12-16 October 2020 |
会议地点 | Seattle, United States |
摘要 | The key to efcient person search is jointly localizing pedestrians and learning discriminative representation for person re-identifcation (re-ID). Some recently developed task-joint models are built with separate detection and re-ID branches on top of shared region feature extraction networks, where the large receptive feld of neurons leads to background information redundancy for the following re-ID task. Our diagnostic analysis indicates the task-joint model suffers from considerable performance drop when the background is replaced or removed. In this work, we propose a subnet to fuse the bounding box features that pooled from multiple ConvNet stages in a bottom-up manner, termed bottom-up fusion (BUF) network. With a few parameters introduced, BUF leverages the multi-level features with different sizes of receptive felds to mitigate the backgroundbias problem. Moreover, the newly introduced segmentation head generates a foreground probability map as guidance for the network to focus on the foreground regions. The resulting foreground attention module (FAM) enhances the foreground features. Extensive experiments on PRW and CUHK-SYSU validate the effectiveness of the proposals. Our Bottom-Up Foreground-Aware Feature Fusion (BUFF) network achieves considerable gains over the state-of-thearts on PRW and competitive performance on CUHK-SYSU. |
收录类别 | EI |
语种 | 英语 |
七大方向——子方向分类 | 图像视频处理与分析 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/44901 |
专题 | 复杂系统认知与决策实验室_智能系统与工程 |
通讯作者 | Huang, Kaiqi |
作者单位 | 1.Institute of Automation, Chinese Academy of Sciences 2.University of Chinese Academy of Sciences 3.Center for Research on Intelligent System and Engineering 4.CAS Center for Excellence in Brain Science and Intelligence Technology |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Yang, Wenjie,Li, Dangwei,Chen, Xiaotang,et al. Botom-Up Foreground-Aware Feature Fusion for Person Search[C],2020. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Bottom-Up Foreground(2206KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
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