Huiping Su
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Huiping Su has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Guang-Bin Huang
School of Automation, Southeast University, Nanjing, China
1 shared publication Hongbo Gao
Department of Automation, School of Information Science and Technology, Institute of Advanced Technology, University of Science and Technology of China, Hefei, China
1 shared publication- Qi Cao
State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
1 shared publication - Qingchao Liu
Automotive Engineering Research Institute, Jiangsu University, Zhenjiang, China
1 shared publication - Xiaozhao Fang
School of Automation, Guangdong University of Technology, Guangzhou, China
1 shared publication Xin Li
School of Computing Science, University of Glasgow, Glasgow, U.K.
1 shared publication
Latest Publications
- 11 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-10
Object Detection in Adverse Weather for Autonomous Vehicles Based on Sensor Fusion and Incremental Learning
Object detection is critical for autonomous vehicles. It enables safe and efficient autonomous driving. However, variations in weather conditions, such as night, fog, rain, and snow, introduce domain shifts that can lead to catastrophic forgetting in detection models, significantly degrading their performance in object detection tasks. This work introduces D-FusionNet, a model that combines data-level sensor fusion with domain incremental learning (Domain-IL) to improve detection performance under such conditions. By combining the complementary strengths of radar and camera sensors, and using Domain-IL to selectively retain features from different weather domains, D-FusionNet demonstrates improved accuracy in object detection, as validated on the RADIATE dataset. This approach effectively addresses catastrophic forgetting and increases object detection accuracy in adverse weather conditions, marking a significant advancement in autonomous vehicle technology.
- 11 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-10
Object Detection in Adverse Weather for Autonomous Vehicles Based on Sensor Fusion and Incremental Learning
Object detection is critical for autonomous vehicles. It enables safe and efficient autonomous driving. However, variations in weather conditions, such as night, fog, rain, and snow, introduce domain shifts that can lead to catastrophic forgetting in detection models, significantly degrading their performance in object detection tasks. This work introduces D-FusionNet, a model that combines data-level sensor fusion with domain incremental learning (Domain-IL) to improve detection performance under such conditions. By combining the complementary strengths of radar and camera sensors, and using Domain-IL to selectively retain features from different weather domains, D-FusionNet demonstrates improved accuracy in object detection, as validated on the RADIATE dataset. This approach effectively addresses catastrophic forgetting and increases object detection accuracy in adverse weather conditions, marking a significant advancement in autonomous vehicle technology.