In this paper, a new photo-based approach for fast human body shape modeling is proposed. This approach takes the user's front and side digital photos as input, and then quickly generates a semblable body shape model with quadrangle mesh. The approach includes five steps: (1) extract profiles from user's photos and divide body contour into parts, (2) automatically select body part from pre-build body parts database, (3) conduct parametric axial deformation for each part, (4) refine body parts based on FFD, (5) smooth the body mesh. The algorithms are elaborated in detail. Experiments show that this approach can achieve good results very quickly and cheaply. © 2010 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Liu, X. H., Wu, Y. W., & Huang, Y. T. (2010). A new photo-based approach for fast human body shape modeling. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6318 LNCS, pp. 119–127). https://doi.org/10.1007/978-3-642-16515-3_16
Mendeley helps you to discover research relevant for your work.