Jiaxin Quan
Papers
2
Total Citations
8
H-Index
2
About
Jiaxin Quan is a researcher specializing in visual object tracking for robotics and computer vision, with a particular focus on improving the robustness of tracking-by-detection systems. Their major contributions center on advancing Multiple Instance Learning (MIL) frameworks to address the critical problem of classifier degradation in online tracking. In their most cited work, "Multiple instance learning tracking based on Fisher linear discriminant with incorporated priors" (2018, 5 citations), Quan introduced a semi-supervised learning model that integrates Fisher linear discriminant analysis with prior knowledge, significantly enhancing tracking stability compared to traditional self-learning methods. Building on this foundation, their subsequent paper "Visual Tracking Using Improved Multiple Instance Learning with Co-training Framework for Moving Robot" (2018, 3 citations) extended the approach to mobile robotics, designing and validating a complete object tracking system that combines improved MIL with a co-training framework. This work directly addresses the practical challenges of enabling natural human-robot interaction through reliable object detection and tracking on moving platforms. While their citation counts reflect a focused, early-career impact, Quan’s contributions are notable for bridging theoretical advances in semi-supervised learning with real-world robotic applications, offering practical solutions to the persistent challenge of maintaining classifier accuracy during autonomous tracking.
Research Focus
Key Achievements
Top Papers
- 1
- 2