Research
Demo Videos
A successful Ph.D. defense with an impressive robot pouring live demo
Yongqiang Huang demonstrated his robotic pouring research result in his Ph.D. defense. The pouring motion was generated using a recurrent neural network.
Drinking Game Just Become Easier with a Beer-Pouring Robot
The demo shows our robotic motion generator can pour any arbitrary amount of liquid accurately from various unseen containers without changing the code. This recurrent neural network based motion generator shows good generalization behavior.
Ideomotor Learning for Robotic Manipulation
Yu Sun's talk at Stanford University during the Bay Area Robotics Symposium 2016
Branching Segment Detection Hamlyn
Example results in our TBME paper "Efficient Vessel Feature Detection for Endoscopic Image Analysis" by Bingxiong Lin, Yu Sun, Jaime E. Sanchez, and Xiaoning Qian
Vessel Tracing Process
The vessel tracing process in our paper "Efficient Vessel Feature Detection for Endoscopic Image Analysis" by Bingxiong Lin, Yu Sun, Jaime E. Sanchez, and Xiaoning Qian
Execution after learning the tasks
Lin, Y., Sun, Y. (2013) Task-Oriented Grasp Planning Based on Disturbance Distribution, ISRR, pp 1-16