General Information
    • ISSN: 1793-8201 (Print), 2972-4511 (Online)
    • Abbreviated Title: Int. J. Comput. Theory Eng.
    • Frequency: Quarterly
    • DOI: 10.7763/IJCTE
    • Editor-in-Chief: Prof. Mehmet Sahinoglu
    • Associate Editor-in-Chief: Assoc. Prof. Alberto Arteta, Assoc. Prof. Engin Maşazade
    • Managing Editor: Ms. Mia Hu
    • Abstracting/Indexing: Scopus (Since 2022), INSPEC (IET), CNKI,  Google Scholar, EBSCO, etc.
    • Average Days from Submission to Acceptance: 192 days
    • E-mail: ijcte@iacsitp.com
    • Journal Metrics:

Editor-in-chief
Prof. Mehmet Sahinoglu
Computer Science Department, Troy University, USA
I'm happy to take on the position of editor in chief of IJCTE. We encourage authors to submit papers concerning any branch of computer theory and engineering.

IJCTE 2021 Vol.13(3): 79-83 ISSN: 1793-8201
DOI: 10.7763/IJCTE.2021.V13.1293

A Novel Active Object Detection Network Based on Historical Scenes and Movements

Nan Ye, Ruogu Wang, and Ning Li

Abstract—Active object detection (AOD) aims at getting a better image for detection. It moves robot to get a best view of the object. This method is more effective than traditional object detection when facing following problems: the object in image is of tiny scale, the object is blocked by other irrelevant objects, the object is partially captured by camera, etc. We consider that the past images acquired by the robot are related to the problems mentioned above. So different from the state-of-the-art methods which mostly generate actions based on current image, a novel AOD network based on CNN and LSTM networks is proposed to advance detection performance in this paper. The AOD network uses current and historical scenes as well as movements are learned to explore and generate following robot actions. We train the Action Network through reinforcement learning. During training, phenomenon that robot stuck and repetitive in several specific situations usually occurred, which results in invalid training. To solve this, an effective training strategy is proposed to skip trapped or repetitive actions. Our proposed AOD network was evaluated on Active Vision Dataset and the experiment results showed its advantage of both accuracy and efficiency on AOD tasks.

Index Terms—Active vision, active object detection, reinforcement learning, convolutional network (CNN), long short-term memory (LSTM).

Nan Ye, Ruogu Wang, and Ning Li are with Shanghai Jiao Tong University, Shanghai, 200240, China (e-mail: Nan-yean@sjtu.edu.cn, rgwang@sjtu.edu.cn, ning_li@sjtu.edu.cn).

[PDF]

Cite:Nan Ye, Ruogu Wang, and Ning Li, "A Novel Active Object Detection Network Based on Historical Scenes and Movements ," International Journal of Computer Theory and Engineering vol. 13, no. 3, pp. 79-83, 2021.

Copyright © 2021 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).


Copyright © 2008-2024. International Association of Computer Science and Information Technology. All rights reserved.