Mengxuan Sun
Papers
1
Total Citations
8
H-Index
1
About
Mengxuan Sun is a researcher advancing the field of 3D computer vision, with a primary focus on category-level 6D object pose estimation—a critical capability for enabling robotic manipulation, augmented reality, and autonomous scene understanding. Sun’s key contributions lie in developing geometry-guided, instance-aware frameworks that allow accurate prediction of an object’s translation and rotation across arbitrary instances within a known category, without requiring per-instance 3D models. Their most-cited work, "Category-Level 6D Pose Estimation Using Geometry-Guided Instance-Aware Prior and Multi-Stage Reconstruction" (2023, 8 citations), introduces a novel approach that leverages geometric priors and multi-stage refinement to overcome challenges in generalization and precision. This work has been recognized for its potential to bridge the gap between controlled lab settings and real-world applications, where objects vary widely in shape and appearance. Sun’s research is particularly impactful for fields like robotics and AR, where robust, model-free pose estimation remains a bottleneck. With a growing citation record, Sun is establishing themselves as a rising voice in 3D perception, pushing the boundaries of how machines understand and interact with the physical world.
Research Focus
Key Achievements
Top Papers
- 1