Title
Combining biomechanical and data-driven body surface models
Abstract
Statistical body shape modelling can be used to realistically generate complex muscle deformation effects on the skin. However, purely data-driven models still ignore the biomechanical nature of surface deformations. Reliable anatomically and biomechanically consistent predictions are barely possible. Our research aims at combining the previously separate paradigms - data-driven and simulation-driven 3D surface modeling - to a hybrid body shape model. Our first goal consists of synthesizing the skin surface from simulated biomechanical data. As a first step in this direction we show preliminary results of our model of an elbow flexion motion with separate biceps and triceps muscle bulging that exhibits believable muscular deformation effects on the skin surface while enabling singular control over specific muscle regions. Our model is separately controllable in shape and pose and extensible to a wider range of human body shapes, joint motion and muscle regions.
Year
DOI
Venue
2017
10.1145/3102163.3102169
SIGGRAPH Posters
Field
DocType
ISBN
Computer vision,Biceps,Elbow,Data-driven,Singular control,Computer science,Body shape,Artificial intelligence,Deformation (mechanics),Triceps Muscle
Conference
978-1-4503-5015-0
Citations 
PageRank 
References 
0
0.34
4
Authors
3
Name
Order
Citations
PageRank
Stefanie Gassel100.34
Thomas Neumann22523156.50
Markus Wacker31239.33