Human Motion Recovery
Egocentric Full-Body Pose Recovery
A synchronized wearable capture and multi-view reconstruction platform for learning full-body pose from head- and wrist-mounted fisheye cameras.
Robotics Researcher
Haoran (Cris) Wang
Hi, I am Haoran, an M.S. student in Mechanical Engineering at Johns Hopkins University. I received my B.Eng. in Robotics Engineering from the Southern University of Science and Technology.
My research focuses on human motion recovery, teleoperation, and multi-sensor fusion. I am also interested in robot manipulation and dexterous manipulation.
Selected work
Research
Multi-Sensor Fusion
Self-Calibration for Resistive Insole Pressure Sensors
An optimization-based calibration method for wearable pressure sensing that reduces vertical ground-reaction-force error.
Robot Design
SWhegPro3 Wheel-Leg Transformable Robot
A four-module mobile robot with transformable three-impeller wheels for robust stair climbing across varied dimensions.
Projects
Humanoid Teleoperation and Motion Retargeting
A modular system that unifies PICO, Xsens, and camera-based human motion for humanoid retargeting, replay, and robot deployment.
Persistent Object-Centric 3D Memory
An RGB-D perception pipeline that combines geometry-based segmentation, SAM2 refinement, and world-frame 3D object representations for persistent scene memory.