A Visual Place Recognition Approach using Learnable Feature Map Filtering and Graph Attention Networks

Cao Qin, Yunzhou Zhang, Yingda Liu, Sonya Coleman, Huijie Du, Dermot Kerr

Research output: Contribution to journalArticlepeer-review


Visual place recognition (VPR) in environments subject to extreme appearance variation due to changing weather, illumination or seasons is a challenging task. Recent works have shown that features learned from CNNs can achieve promising performance. However, most of the existing methods concentrate so much on the image itself that they neglect the architecture of the network, especially dierent lters that may carry more meaningful information. In this paper, we develop a learnable feature map filtering (FMF) module constrained by triplet loss to re-calibrate the weight of the individual feature map. In this way, specic feature maps that encode invariant characteristics of location are extracted. Moreover, to make full use of the
rich global mutual information that resides in the sample set, we propose an influence-based graph attention network (IB-GAT) with a verication subnet to better incorporate the relations among samples during the training process. Dierent from conventional GAT approaches, IB-GAT enables feature nodes to attend over
the influence of other nodes instead of the original feature. Thus rened features with more discriminative power could be generated. Extensive experiments have been conducted on six public VPR datasets with varying appearances. Ablation analysis verifies the potential ecacy of the FMF module and the IB-GAT components. The experimental results also demonstrate that the proposed methods can achieve better performance than the current state of the art.
Original languageEnglish
Article numberNEUCOM-D-20-00767R1
Number of pages19
Publication statusAccepted/In press - 20 Feb 2021


  • Visual place recognition
  • Convolutional Neural Networks
  • Feature Map Filtering
  • Graph Attention Networks


Dive into the research topics of 'A Visual Place Recognition Approach using Learnable Feature Map Filtering and Graph Attention Networks'. Together they form a unique fingerprint.

Cite this