Automatic shape modeling of the foot: towards a database of foot shapes
Homologous human shape modeling is critical to utilize 3D human body data for product design. Software libraries for homologous modeling were developed and application software to model the foot form was also developed using the libraries. This software w
- PDF / 794,814 Bytes
- 6 Pages / 481.89 x 685.984 pts Page_size
- 94 Downloads / 134 Views
582
Automatic shape modeling of the foot: towards a database of foot shapes Masaaki Mochimarua,b, Makiko Kouchia,b a Digital Human Lab., National Institute of Advanced Industrial Science and Technology, 2-41-6 Aomi, Koto-ku, Tokyo 135-0064 JAPAN b CREST, JST (Japan Science and Technology Corporation)
Abstract Homologous human shape modeling is critical to utilize 3D human body data for product design. Software libraries for homologous modeling were developed and application software to model the foot form was also developed using the libraries. This software was incorporated into a compact low-cost foot scanner. With the scanning system, retailers can select/recommend well-fitting shoes to customers. By storing a copy of scanned data that does not contradict with the privacy of customers, customers can enjoy browsing their foot shape data or shopping in electronic commerce. This system is useful to make a large database of foot forms. Possible problems in making a large database in this manner are discussed. Keywords: Anthropometry, human body shape, foot
1. Introduction With 3D body scanners, a large database on body dimensions and forms can be developed efficiently. It is difficult, however, to utilize the information on the 3D human body forms for the product design only with millions of surface data points and dozens of anatomical landmarks. Millions data points are verbose to describe a body form, and they have no anatomical correspondence. Landmarks have anatomical correspondence, but they are too few to represent the surface form. Thus, shape representation methods using hundreds to thousands data points with anatomical correspondence have been proposed. Such methods are called "normalized modeling" [I] or "homologous shape modeling" [2]. The modeling protocol is not unique, but the modeling guideline is defined reasonably based on anatomical correspondence and application purposes of the models. In this paper, the common and basic protocols for the homologous modeling are mentioned, and development of a foot scanning system with an automatic shape modeling software is mentioned. Finally, a method is proposed for collecting a large amount of foot shape data using the foot scanning systems installed in shoe retailers.
CARS 2002 - H. U. Lemke, M. W. Vannier; K. Inamura, A.C. Farman, K. Doi & J.H.c. Reiber (Editors) ' CARS/Springer. All rights reserved.
583
2. Homologous shape modeling 2.1 Basic modules designed for homologous shape modeling We have developed several homologous shape models for different body parts. Fig. l(a) shows a torso model for developing a dressmaking dummy [3]. It consists of 547 data points (1039 polygons) defined based on 25 landmarks. Fig. I (b) is a face model for designing a spectacle frame, which consists of 211 data points (366 polygons) based on 61 anatomical landmarks [4]. A foot model for shoe last design is shown in Fig. I(c). This model has 295 data points (586 polygons) defined based on 9 landmarks [2]. Common and basic protocols for the geometrical processing to gene
Data Loading...
