Model-Based Generation of Realistic 3D Full Body Avatars from Uncalibrated Multi-view Photographs

Abstract : In today’s world of rapid technological advancement, we find an increasing demand for low-cost systems that are capable of fast and easy generation of realistic avatars for use in Virtual Reality (VR) applications. For example, avatars can enhance the immersion experience of users in video games and facilitate education in virtual classrooms. Therefore, we present here a novel model-based technique that is capable of real-time generation of personalized full-body 3D avatars from orthogonal photographs. The proposed method utilizes a statistical model of human 3D shape and a multi-view statistical 2D shape model of its corresponding silhouettes. Our technique is automatic, requiring minimal user intervention, and does not need a calibrated camera. Each component of our proposed technique is extensively evaluated and validated.