CEKD: Cross-Modal Edge-Privileged Knowledge Distillation for Semantic Scene Understanding Using Only Thermal Images

Zhen Feng, Yanning Guo, Yuxiang Sun

Research output: Journal article publicationJournal articleAcademic researchpeer-review

14 Citations (Scopus)

Abstract

Semantic scene understanding using thermal images has received great attention due to the advantage that thermal imaging cameras could see in challenging illumination conditions. However, thermal images are lack of color information and the edges in thermal images are often blurred, making them not very suitable to be directly used by existing semantic segmentation networks that are designed with RGB images. To address this problem, we propose a cross-modal edge-privileged knowledge distillation framework, which utilizes a well-trained RGB-Thermal fusion-based semantic segmentation network with edge-privileged information as the teacher, to guide the training of a semantic segmentation network as the student. The student network only uses thermal images. The experimental results on a public dataset demonstrate that under the guidance of the teacher, the student network achieves superior performance over the state of the arts using only thermal images.

Original languageEnglish
Pages (from-to)2205-2212
Number of pages8
JournalIEEE Robotics and Automation Letters
Volume8
Issue number4
DOIs
Publication statusPublished - 1 Apr 2023

Keywords

  • autonomous driving
  • Knowledge distillation
  • privileged information
  • semantic segmentation
  • thermal images

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Biomedical Engineering
  • Human-Computer Interaction
  • Mechanical Engineering
  • Computer Vision and Pattern Recognition
  • Computer Science Applications
  • Control and Optimization
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'CEKD: Cross-Modal Edge-Privileged Knowledge Distillation for Semantic Scene Understanding Using Only Thermal Images'. Together they form a unique fingerprint.

Cite this