A Statistical Model of Human Pose and Body Shape

Abstract

Generation and animation of realistic humans is an essential part of many projects in today's media industry. Especially, the games and special effects industry heavily depend on realistic human animation. In this work a unified model that describes both, human pose and body shape is introduced which allows us to accurately model muscle deformations not only as a function of pose but also dependent on the physique of the subject. Coupled with the model's ability to generate arbitrary human body shapes, it severely simplifies the generation of highly realistic character animations. A learning based approach is trained on approximately 550 full body 3D laser scans taken of 114 subjects. Scan registration is performed using a non-rigid deformation technique. Then, a rotation invariant encoding of the acquired exemplars permits the computation of a statistical model that simultaneously encodes pose and body shape. Finally, morphing or generating meshes according to several constraints simultaneously can be achieved by training semantically meaningful regressors.

Posture-invariant statistical shape analysis using Laplace operator

“…Xi et al [20] improve this approach by adding a step transforming the template model using a radial basis function based on the known marker positions. Hasler et al [10] extend Allen et al's approach to work for models in varying postures. As before, a set of labeled marker positions is required for this method.…”

“…In our experiments, we use two databases: the CAESAR database [1] and the MPI database [10]. The MPI database contains 520 models of 111 different individuals in up to 35 different postures.…”

“…They apply the multilinear analysis to sets of facial images that combine different facial geometries, expressions, head poses, and lighting conditions. Allen et al [4], Anguelov et al [9], and Hasler et al [10] propose statistical models to encode human body shape. Allen et al only encode the shape variation in a standard posture.…”

“…The MPI database contains 520 models of 111 different individuals in up to 35 different postures. The models are parameterized using the approach by Hasler et al [10] and each parameterized model contains 6449 vertices. The markers used for parameterization were placed manually on the scans.…”

Towards Plenoptic Raumzeit Reconstruction

“…Given the input images a 3D representation for each discrete time step is reconstructed and used for depth guided resampling of the plenoptic function [17][18][19][20][21][22]. For restricted scene setups the incorporation of template models proves beneficial [23][24][25]. Only few approaches reconstruct a temporally consistent mesh, which allows for continuous time interpolation [26,27].…”