11am - 12 noon
Thursday 13 February 2020
An affordable real-time volumetric capture system
Careful camera interfaces, controlled environments and further processing will significantly improve the quality of the models and meet requirements for the purposes.
University of Surrey
This event has passed
This seminar introduces how to build a real-time volumetric capture system using multiple commodity depth cameras and reliable open sources. The system includes camera synchronization and calibration, volumetric reconstruction, mesh generation, and texture mapping. This system provides the base line of real-time multi-view 3D reconstruction with visually acceptable qualities. Thus it will not be suitable for high-quality holographic productions, but be potential for real-time experiments in uncontrolled spaces with cheap costs.
Careful camera interfaces, controlled environments, and further processing will significantly improve the quality of the models and meet requirements for the purposes. Finally I will summarize my visiting research work here regarding dynamic human modeling in the multi-view environment.
Hwasup Lim received his PhD degree in electrical engineering from the Pennsylvania State University in 2007. He was a R&D staff member at the Advanced Media Lab, Samsung Advanced Institute of Technology
(SAIT) from 2007 to 2011, where his research focused on TOF depth image processing, and 3D human modeling and tracking. He is currently a principal research scientist at the Center for Imaging Media Research, Korea Institute of Science and Technology (KIST). His research interests include RGBD image-based human modeling and tracking.