Weakly-Supervised Object Localization by Cutting Background with Deep Reinforcement Learning | |
Wu Zheng1,2,4; Zhaoxiang Zhang1,2,3,4 | |
2018-08 | |
会议名称 | Pacific Rim International Conference on Artificial Intelligence 2018 |
会议录名称 | Lecture Notes in Computer Science |
卷号 | vol 11013 |
会议日期 | 2018-08 |
会议地点 | Nanjing, China |
出版地 | Cham |
出版者 | Springer |
摘要 | Weakly-supervised object localization only depends on image-level labels to obtain object locations and attracts more attention recently. Taking inspiration from the human visual mechanism that human searches and localizes the region of interest by shrinking the view from a wide range and ignoring the unrelated background gradually, we propose a novel weakly-supervised localization method of cutting background of an object iteratively to achieve object localization with deep reinforcement learning. This approach can train an agent as a detector, which searches through the image and tries to cut off all regions unrelated to classification performance. An effective refinement approach is also proposed, which generates a heat-map by sum-pooling all feature maps to refine the location cropped by the agent. As a result, by combining the top-down cutting process and the bottom-up evidence for refinement, we can achieve a good performance on object localization in only several steps. To the best of our knowledge, this may be the first attempt to apply deep reinforcement learning to weakly-supervised object localization. We perform our experiments on PASCAL VOC dataset and the results show our method is effective. |
关键词 | Weakly-supervised Object Localization, Deep Reinforcement Learning, Convolutional Neural Network |
学科门类 | 模式识别与计算机视觉 |
DOI | https://doi.org/10.1007/978-3-319-97310-4_24 |
URL | 查看原文 |
收录类别 | EI |
语种 | 英语 |
引用统计 | |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/23860 |
专题 | 智能感知与计算研究中心 |
通讯作者 | Zhaoxiang Zhang |
作者单位 | 1.Center for Research on Intelligent Perception and Computing, CASIA, China 2.National Laboratory of Pattern Recognition, CASIA, China 3.CAS Center for Excellence in Brain Science and Intelligence Technology, China 4.University of Chinese Academy of Sciences, China |
第一作者单位 | 模式识别国家重点实验室; 中国科学院自动化研究所 |
通讯作者单位 | 模式识别国家重点实验室; 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Wu Zheng,Zhaoxiang Zhang. Weakly-Supervised Object Localization by Cutting Background with Deep Reinforcement Learning[C]. Cham:Springer,2018. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
PRICAI_2018_ID188_We(1765KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
个性服务 |
推荐该条目 |
保存到收藏夹 |
查看访问统计 |
导出为Endnote文件 |
谷歌学术 |
谷歌学术中相似的文章 |
[Wu Zheng]的文章 |
[Zhaoxiang Zhang]的文章 |
百度学术 |
百度学术中相似的文章 |
[Wu Zheng]的文章 |
[Zhaoxiang Zhang]的文章 |
必应学术 |
必应学术中相似的文章 |
[Wu Zheng]的文章 |
[Zhaoxiang Zhang]的文章 |
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