Peiyang Shi
Papers
2
Total Citations
6
H-Index
2
About
Peiyang Shi’s research lies at the intersection of robotics, computer vision, and symbolic artificial intelligence, with a focus on enabling machines to perceive and manipulate complex, unstructured environments. His most notable contribution addresses the long-standing challenge of robotic manipulation of deformable objects—such as cloth, cables, or soft materials—which lack standard geometric representations. In his 2020 work, Shi introduced sequential topological representations that compactly capture the state of highly deformable objects, enabling predictive models to anticipate their behavior. This innovation provides a principled framework for robots to reason about and control objects that were previously intractable for autonomous systems. Additionally, Shi has explored unsupervised object detection for symbolic representation, bridging deep learning with classical AI to extract meaningful symbols from raw visual data. Though his citation counts are still building—with 4 and 2 citations respectively—the conceptual novelty of his work has positioned him as an emerging voice in deformable object manipulation. His research offers a promising path toward more adaptable robots capable of handling the messy, non-rigid realities of the physical world.
Research Focus
Key Achievements
Top Papers
- 1
- 2Faster Unsupervised Object Detection For Symbolic Representation2 citations · 2020