Fulin Liu
Papers
1
Total Citations
19
H-Index
1
About
Fulin Liu is a rising researcher in computer vision and robotics, specializing in 6D object pose estimation—a critical capability for augmented reality and robotic manipulation. His most influential work, "NVR-Net: Normal Vector Guided Regression Network for Disentangled 6D Pose Estimation" (2023), addresses a fundamental challenge in monocular pose estimation: the entanglement of rotation and translation in traditional PnP-based methods. Liu introduced a novel normal vector-guided regression framework that decouples these components, achieving superior accuracy by predicting rotations independently from translations. This approach has garnered 19 citations since publication, signaling strong interest from the community. Liu's contribution lies in rethinking how neural networks can bypass geometric bottlenecks, offering a cleaner, more robust solution for real-world applications. His work exemplifies a trend toward disentangled representations in 3D vision, and NVR-Net has been recognized for its practical impact on autonomous systems. As an early-career researcher, Liu is establishing himself as a thoughtful innovator in pose estimation, with potential to influence both algorithmic design and deployment in robotics.
Research Focus
Key Achievements
Top Papers
- 1