Full-body Modeling and Tracking for Embodiment in VR Dark Rides

Real-time full-body tracking based on single or multiple commodity depth sensors are an integral part of many applications such as motion transfer in 3D games, 3D virtual and augmented reality, and human computer interaction. The proposed project directly derives from practical needs expressed by the partner organization, DreamCraft Attractions (Victoria) in developing fully immersive VR dark rides (i.e. for amusement parks). Our research targets highly robust and low-latency full-body motion capture in the presence of interfering geometry such as chairs or props used in VR dark rides, using a multi-view camera system. In particular, the proposed project would adapt articulated tracking algorithms developed in [Tagliasacchi et al. 2015; Tkach et al. 2016] to build a highly robust, accurate and efficient prototype for full-body motion tracking, while addressing the research challenges specific to this application domain. TO BE CONT’D

Faculty Supervisor:

Andrea Tagliasacchi

Student:

Jiju Poovvancheri

Partner:

DreamCraft Attractions Ltd

Discipline:

Computer science

Sector:

Media and communications

University:

Program: