Shimeng Fan
Papers
2
Total Citations
49
H-Index
2
About
Shimeng Fan is a leading researcher in computer vision, with a primary focus on 6-degree-of-freedom (6D) object pose estimation and tracking—a critical technology for applications in multimedia, augmented reality, and robotic manipulation. Fan’s most influential work, the HFF6D framework (2022, 41 citations), introduces a hierarchical feature fusion network that robustly tracks object pose in challenging video sequences, overcoming common failures such as incorrect initial poses, sudden re-orientation, and severe occlusions. This contribution significantly advances the reliability of 6D tracking in real-world, dynamic environments. More recently, Fan has pushed the boundaries of category-level pose estimation through a novel approach combining fine-grained segmentation with difference-aware shape adjustment (2023, 8 citations), enabling more accurate and adaptable pose inference across diverse object instances without requiring CAD models. By addressing both instance-level and category-level challenges, Fan’s work bridges a crucial gap between theoretical pose estimation and practical deployment. Their research is widely recognized for its impact on robust, real-time vision systems, making Fan a key figure in the ongoing evolution of intelligent robotic and multimedia interaction.
Research Focus
Key Achievements
Top Papers
- 1
- 2