Introduction to the Special Issue on Representation, Analysis, and Recognition of 3D Humans

Modeling, processing, recognizing, searching, and retrieving 3D human data (shapes, gestures, interactions) is a well-established research area in Multimedia. In the last decade, there has been a tremendous increase in opportunities for using 3D human data in medicine, security, and human computer interaction, largely driven by the development of effective devices and algorithms for recovering 3D data (e.g., Microsoft Kinect, Intel RealSense, Google Project Tango, and Apple Prime-Sense). Such rich information opens the way to new modes of experiential computing, interactive environments, as well as new multimedia content. This special issue is of interest to an interdisciplinary target audience as well interdisciplinary teams of contributors spanning: applied math, multimedia experiential computing, computational science and engineering, and application domain experts. Several fundamental research problems within the scope include: Representations for 3D static and dynamic human data Representations for non-rigid 3D objects (face, body) Temporal modeling of 3D face/body sequences Machine learning techniques for 3D human representations Computationally efficient strategies for resource constrained deployments Fusing multiple cues: shape, color, texture, motion etc. This special issue aims to bring together researchers interested in defining new and innovative solutions that advance the way 3D human data are used in multimedia computing, communications and applications such as human behavior understanding from 3D sensors, animation and entertainment, sports analytics, natural interaction, virtual and augmented reality. Application areas of interest include, but are not limited to: