Papers
2
Total Citations
14
H-Index
2
About
Xin Shu is a robotics researcher whose work focuses on the intersection of computer vision and tactile sensing to advance autonomous robotic manipulation. Their key research areas include instance segmentation for grasping, tactile-based stability prediction, and deep learning for robotic control. Shu’s major contributions include developing a self-supervised learning approach that leverages instance segmentation to improve robotic grasping in cluttered and occlusion-heavy environments, addressing a critical limitation of traditional methods that struggle with texture-less objects. This work, published in 2018, has garnered 9 citations and laid the groundwork for more robust grasping in novel settings. More recently, Shu proposed a novel convolutional neural network architecture that fuses and reconstructs tactile sensor data to predict grasping stability across objects of varying shapes. This 2022 study, with 5 citations, demonstrates a significant advance in using tactile feedback—rather than just visual input—to ensure reliable grasps. By combining visual and tactile modalities, Shu’s research is paving the way for more dexterous and adaptable robots capable of handling real-world uncertainty. Their work is particularly influential for students and researchers interested in sensor fusion, deep learning for robotics, and autonomous manipulation.
Research Focus
Key Achievements
Top Papers
- 1
- 2