Journal of Systems Engineering and Electronics ›› 2023, Vol. 34 ›› Issue (2): 460-468.doi: 10.23919/JSEE.2023.000050


Relational graph location network for multi-view image localization

Yukun YANG(), Xiangdong LIU()   

  1. 1 School of Automation, Beijing Institute of Technology, Beijing 100081, China
  • Received:2022-01-04 Online:2023-04-18 Published:2023-04-18
  • Contact: Yukun YANG;
  • About author:
    YANG Yukun was born in 1991. She received her Ph.D. degree in control science and engineering from Beijing Institute of Technology, China. Her research interests include image/video recognition and generation, image localization and transfer learning. E-mail:

    LIU Xiangdong was born in 1971. He received his Ph.D. degree in space engineering from Harbin Institute of Technology, China. He is currently a professor with the School of Automation, Beijing Institute of Technology, China. His research interests include high-precision servo control, motor drive control, piezoceramics actuator drive and compensation control, sliding model control, state estimation, and attitude control. E-mail:


In multi-view image localization task, the features of the images captured from different views should be fused properly. This paper considers the classification-based image localization problem. We propose the relational graph location network (RGLN) to perform this task. In this network, we propose a heterogeneous graph construction approach for graph classification tasks, which aims to describe the location in a more appropriate way, thereby improving the expression ability of the location representation module. Experiments show that the expression ability of the proposed graph construction approach outperforms the compared methods by a large margin. In addition, the proposed localization method outperforms the compared localization methods by around 1.7% in terms of meter-level accuracy.

Key words: multi-view image localization, graph construction, heterogeneous graph, graph neural network