Structure-aware Feature Disentanglement with Knowledge Transfer for Appearance-changing Place Recognition

Cao Qin, Yunzhou Zhang, Yingda Liu, Delong Zhu, Sonya Coleman, Dermot Kerr

Research output: Contribution to journalArticlepeer-review

Abstract

Long-term visual place recognition is challenging as the environment is subject to drastic appearance changes across different temporal resolutions such as time of the day, month, and season. A wide variety of existing methods address the problem by means of feature disentangling or image style transfer, but ignore the structural information which often remains stable even under environmental condition changes. To overcome this limitation, this paper presents a novel Structure-aware Feature Disentanglement Network (SF-DNet) based on knowledge transfer and adversarial learning. Explicitly, probabilistic knowledge transfer (PKT) is employed to transfer knowledge obtained from the Canny edge detector to the structure encoder. An appearance teacher module is then designed to ensure the learning of appearance encoder does not only rely on metric learning. The generated content features with structural information are used to measure the similarity of images. We finally evaluate the proposed approach, and compare it to state-of-the-art place recognition methods using six datasets with extreme environmental changes. Experimental results demonstrate the effectiveness and improvements achieved using the proposed framework.
Original languageEnglish
JournalIEEE Transactions on Neural Networks and Learning Systems
Publication statusAccepted/In press - 11 Aug 2021

Keywords

  • Disentanglement
  • Representation
  • Changing environment
  • Knowledge transfer
  • Visual place recognition

Fingerprint

Dive into the research topics of 'Structure-aware Feature Disentanglement with Knowledge Transfer for Appearance-changing Place Recognition'. Together they form a unique fingerprint.

Cite this