Active Object Detection Using Double DQN and Prioritized Experience Replay
文献类型：会议论文
| 作者 | Liu H(刘华平)1; Yang, Dongfang3; Sun FC(孙富春)1; Han XN(韩小宁)2
|
| 出版日期 | 2018 |
| 会议日期 | July 8-13, 2018 |
| 会议地点 | Rio de Janeiro, Brazil |
| 页码 | 1-7 |
| 英文摘要 | Visual object detection is one of the fundamental tasks in computer vision and robotics. Small scale, partial capture and occlusion often occur in robotic applications, most existing object detection algorithms perform poorly in such situations. While a robot can look at one object from different views and plan its trajectory in the next few steps, which can lead to better observations. We formulate it as a sequential action-decision process, and develop a deep reinforcement learning architecture to solve the active object detection problem. A double deep Q-learning network (DQN) is applied to predict an action at each step. Experimental validation on the Active Vision Dataset shows the efficiency of the proposed method. |
| 产权排序 | 1 |
| 会议录 | Proceedings of the International Joint Conference on Neural Networks
 |
| 会议录出版者 | IEEE |
| 会议录出版地 | New York |
| 语种 | 英语 |
| ISBN号 | 978-1-5090-6014-6 |
| 源URL | [http://ir.sia.cn/handle/173321/23592]  |
| 专题 | 沈阳自动化研究所_空间自动化技术研究室 |
| 通讯作者 | Han XN(韩小宁) |
| 作者单位 | 1.Department of Computer Science and Technology, Tsinghua University, Beijing, China 2.Chinese Academy of Sciences, Shenyang Institute of Automation, Shenyang, China 3.Xi'An High Tech Research Institution, China |
| 推荐引用方式 GB/T 7714 | Liu H,Yang, Dongfang,Sun FC,et al. Active Object Detection Using Double DQN and Prioritized Experience Replay[C]. 见:. Rio de Janeiro, Brazil. July 8-13, 2018. |
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。