More about HKUST
Reconstructing hands and manipulated objects with graph neural networks
Speaker:Elden TSE School of Computer Science University of Birmingham Title: "Reconstructing hands and manipulated objects with graph neural networks" Date: Monday, 26 September 2022 Time: 4:00pm - 5:00pm Venue: Lecture Theater F (Leung Yat Sing Lecture Theater) (near lift 25/26, HKUST) Abstract: Development of RGB-D sensors have opened a range of technologies and applications which requires hand detections and 3D pose estimations in a variety of scenarios such as AR/VR, robot grasping and assisted driving. However, most existing methods do not generalise well to real-world scenarios and prone to fail when extended to hand-object interactions. In this talk, I will share about my works and highlight the recent works in the field. ****************** Biography: Elden is a PhD student in Computer Science at University of Birmingham. His current research projects focus on hand and object interactions with graph neural networks. His long-term goal is to enable artificial intelligent agents to better understand and interact with the physical world.