Shun‐Cheng Wu

Papers

Total Citations

H-Index

About

Shun-Cheng Wu is a rising star in 3D computer vision and robotics, whose work is fundamentally advancing how machines perceive and interact with the physical world. His research focuses on two critical, intertwined challenges: 6-DoF (six degrees of freedom) robotic grasping and category-level 6D object pose estimation. Wu’s major contributions are marked by a pragmatic shift from instance-specific methods to robust, generalizable systems. His first-author work, **MonoGraspNet** (2023, 42 citations), tackles the long-standing problem of robotic grasping from a single RGB image, demonstrating superior performance over depth-dependent methods, especially on photometrically challenging objects. This work highlights his ability to solve practical, sensor-constrained problems. Perhaps his most significant achievement is the creation of **HouseCat6D** (2024, 26 citations), a large-scale, multi-modal dataset for category-level pose estimation. By providing high-quality annotations and diverse, realistic scenarios, this dataset directly addresses a critical bottleneck in the field, enabling researchers to move beyond simple benchmarks and develop more robust perception systems. Wu’s work is not just about incremental improvements; it provides the foundational data and algorithms necessary for the next generation of autonomous robots.