Geometric Vision and Robotics Lab

Geometric Vision and Robotics Lab

See the unseen, shape the shell 洞见未见,塑以躯壳

[ Team ]

Meet the Team

Geometric Vision and Robotics Lab logo
[ About The Team ]

We build intelligent visual systems that bridge embodied AI and clinical medicine.

Leveraging 3D scene understanding, multimodal agents, and geometry-aware reconstruction to advance spatial perception and interaction across surgical scenarios and digital twins.

[ Research Directions ]

Our Research Areas

Multimodal World Model

Multimodal World Model

Enables systems to perceive, interpret, and reason about dynamic scenes from videos and 3D representations.

Read more →
Geometry Estimation and 3D Reconstruction

Geometry Estimation and 3D Reconstruction

Focuses on recovering accurate 3D geometry, depth, and structure from visual data, especially in medical scenarios.

Read more →
Registration

Registration

Studies robust alignment between 2D and 3D data to support precise applications such as surgical navigation.

Read more →
Digital Twins

Digital Twins

Explores building virtual environments and multimodal intelligent systems for simulation and embodied interaction.

Read more →

[ Publications ]

Featured Research Projects

Multimodal World Model

Enables systems to perceive, interpret, and reason about dynamic scenes from videos and 3D representations.

Geometry Estimation and 3D Reconstruction

Focuses on recovering accurate 3D geometry, depth, and structure from visual data, especially in medical scenarios.

Digital Twins

Explores building virtual environments and multimodal intelligent systems for simulation and embodied interaction.