Extraction of feature parameters of threedimensional human body
based on image
DENG Wei-yan1,2, LU Guo-dong1, WANG Jin1, CHEN Long3
1. State Key Laboratory of CAD&CG, Zhejiang University, Hangzhou 310027, China; 2. College of Mechanical and
Electrical Engineering, China Jiliang University,Hangzhou 310018, China;
3. Department of Mechanical Engineering,
University of Shanghai for Science and Technology, Shanghai 200093, China
A technique for feature point extraction of human body based on image was presented to solve the problem of body feature parameter extraction in threedimensional human body modeling. Then each part of the human body in picture was distinguished, and the calculating process was given. First, the collected image was preprocessed, and then the front and side body feature points were identified to get body feature district, finally, the body feature sizes were extracted. The method to calculate the body feature sizes was introduced according to the modeling method and the reference method. Then a new method was proposed to improve the accuracy of body feature sizes. Human body feature parameters were obtained from the collected images by the modeling reference method. Results show that the method can accurately extract the feature points and the feature sizes.
DENG Wei-Yan, LIU Guo-Dong, WANG Jin, CHEN Long. Extraction of feature parameters of threedimensional human body
based on image. J4, 2010, 44(5): 837-840.
[1] HILTON A, BERESFORD D, GENTILS T, et al. Wholebody modeling of people from multiview images to populate virtual worlds [J]. Visual Computer, 2000, 16(7): 411436.
[2] LEE W, GU J, MAGNENATTHALMANN N. Generating animatable threedimensional virtual humans from photographs [J]. Computer Graphics Forum, 2000, 19(3): 110.
[3] WANG C C L, WANG Y, CHANG T K K, et al. Virtual human modeling from photographs for garment industry [J]. ComputerAided Design, 2003, 35(6): 577589.