Yunhui Guo
Papers
2
Total Citations
12
H-Index
2
About
Yunhui Guo is a roboticist and computer vision researcher whose work sits at the intersection of autonomous manipulation and perception. His primary research focuses on enabling robots to interact with and understand novel objects in unstructured environments, with key contributions in **unseen object instance segmentation** and **novel instance detection**. In his highly cited 2023 work, *"Self-Supervised Unseen Object Instance Segmentation via Long-Term Robot Interaction"* (10 citations), Guo introduced a pioneering robotic system that leverages prolonged physical interaction—pushing and grasping—to self-supervise the segmentation of unfamiliar objects, moving beyond single-action methods to build robust, real-world understanding. More recently, his 2025 paper *"Adapting Pre-Trained Vision Models for Novel Instance Detection and Segmentation"* (2 citations) proposes NIDS-Net, a unified framework that adapts pre-trained vision models to detect and segment novel object instances from just a few examples, blending object proposal generation with embedding learning. Guo’s work is notable for bridging the gap between static perception and interactive learning, demonstrating how robots can use their own actions to generate training data. His research holds significant promise for applications in warehouse automation, home robotics, and any domain requiring flexible object handling.
Research Focus
Key Achievements
Top Papers
- 1
- 2