3D-Articulated Hand Pose Estimation for Multi-touchless Human Machine Interaction

The College of Forestry, Wildlife and Tourism (CFWT) invites SUA community and all intrested people to attend the seminar entitled “3D-Articulated Hand Pose Estimation for Multi-touchless Human Machine Interaction”

This seminar presents an approach to developing a new interface that offers 3D interaction without direct contact with the computing device. This novel interaction technology is spurred by the 3D-hand pose estimation from depth which is a very challenging problem in the computer vision domain. 3D-Hand pose estimation provides a possibility for the new multi-touchless interface leading towards seamless human-machine interaction.

This is important as the conventional computing is limited to 2D plane display and is only suited when users sit behind the computing device. However, there are some scenarios that are inconvenience for a user to sit behind the computing device, for example, when the user is a medical doctor performing surgery (remote surgery), during road traffic interaction, in piloting, during presentation, or in communication with the disabled people (sign language recognition).

To demonstrate the performance of 3D-hand pose estimation in facilitating collaborative computing, a complete experiment was conducted on three challenging public datasets, ICVL, MSRA, and NYU. The empirical results show that 3D-hand pose estimation provides a natural way of interaction between human and virtual space that achieves greater user experience.

 

Speaker: Dr Jamal Banzi (PhD), Information and Communication Engineer, Department of Tourism and Recreation, SUA

Date & time: Tuesday 28^th November, 2019 from 10:00 to 12:00 noon

Venue: National Carbon Monitoring Centre