Yifan Pu
Papers
1
Total Citations
22
H-Index
1
About
Yifan Pu is a rising researcher in robotics and computer vision, with a focused expertise on the challenging problem of manipulating deformable linear objects (DLOs)—such as ropes, cables, and wires. Their most-cited work, "Learning to Estimate 3-D States of Deformable Linear Objects from Single-Frame Occluded Point Clouds" (2023, 22 citations), addresses a critical bottleneck in robotic manipulation: accurately inferring the 3D state of DLOs from noisy, partially occluded sensor data. Pu’s key contribution lies in developing a learning-based framework that robustly estimates the high-dimensional state space of these objects from just a single point cloud frame, overcoming the limitations of traditional methods that struggle with frequent occlusions and sensor noise. This work is foundational for enabling precise, real-world robotic tasks like automated wire harnessing, cable routing, and surgical thread manipulation. By tackling the core issue of state estimation under uncertainty, Yifan Pu is paving the way for more dexterous and autonomous robotic systems capable of handling the complex, flexible objects that are ubiquitous in manufacturing, healthcare, and domestic environments.
Research Focus
Key Achievements
Top Papers
- 1