Xinyue Zhao
Papers
6
Total Citations
119
H-Index
5
About
Xinyue Zhao is a robotics and computer vision researcher whose work centers on industrial automation, visual servoing, and 6D pose estimation for challenging manufacturing environments. Her research has made significant strides in solving one of the field's most persistent problems: accurately detecting and grasping textureless, reflective metal components — such as bearings and flanges — that defeat conventional feature-based methods. Zhao's most influential contribution, "Moment-Based 2.5-D Visual Servoing for Textureless Planar Part Grasping" (2018, 47 citations), addressed critical shortcomings in industrial visual servoing by reformulating moment-based approaches to be more robust and practically deployable. Her subsequent work introduced generative and deep learning frameworks — including ContourPose (2023, 20 citations) and a feature-to-image robotic vision system (2021, 15 citations) — that push the boundaries of monocular 6D pose estimation for reflective surfaces. Beyond manipulation, Zhao has also contributed to service robotics, developing a novel slot-gated model for natural language understanding in task-request scenarios (2019, 12 citations). Collectively, her publications reflect a researcher bridging the gap between theoretical computer vision and real-world industrial robot deployment, with growing recognition across the robotics community.
Research Focus
Key Achievements
Top Papers
- 1Moment-Based 2.5-D Visual Servoing for Textureless Planar Part Grasping47 citations · 2018
- 2A circular feature-based pose measurement method for metal part grasping21 citations · 2017
- 3
- 4
- 5
- 6