Remodeling of mannequins based on automatic binding of mesh to anthropometric parameters

In order to improve the accuracy of semantic anthropometric parameters in expressing the body shape of the mannequin and simplify the manual pre-processing of the parameter to body shape mapping, a method for creating 3D mannequins based on automatic binding of mesh to anthropometric parameters is p...

Full description

Saved in:
Bibliographic Details
Published in:The Visual computer Vol. 39; no. 12; pp. 6435 - 6458
Main Authors: Li, Xihang, Li, Guiqin, Li, Tiancai, Lv, Jianping, Mitrouchev, Peter
Format: Journal Article
Language:English
Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.12.2023
Springer Nature B.V
Subjects:
Animation
Anthropometry
Artificial Intelligence
Binding
Computer Graphics
Computer Science
Customization
Data processing
Datasets
Finite element method
Human body
Image Processing and Computer Vision
Iterative methods
Mannequins
Mapping
Methods
Optimization
Original Article
Parameter estimation
Semantics
Torso
Mannequin modeling
XGBoost-RFECV
Semantic anthropometric parameters
Mesh deformation
Constrained optimization
ISSN:0178-2789, 1432-2315
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In order to improve the accuracy of semantic anthropometric parameters in expressing the body shape of the mannequin and simplify the manual pre-processing of the parameter to body shape mapping, a method for creating 3D mannequins based on automatic binding of mesh to anthropometric parameters is presented. For this purpose, 27 anthropometric parameters are used to accurately constrain the shape of the human body. The missing parameters are estimated by partial user input, which avoids complex input. A method based on eXtreme Gradient Boosting and Recursive Feature Elimination with Cross-Validation to automatically bind the most relevant anthropometric parameters for each mesh is proposed and 3D mannequin is created by local mapping. This makes it easier to add or subtract parameters to constrain the mannequin. An error-based approach for iterative optimization of mannequins is also proposed to ensure that the created mannequin matches the user input parameters and has good local deformation capability. Width and thickness parameters have been added for the torso position of the mannequin, which provides good scalability for creating a mannequin based on 2D anthropometric measurements. The accuracy, flexibility and reliability of the reconstructed method are analyzed through series of performed experiments. An innovative application of personalized mannequin customization under clothing combining Kinect and multi-sensor information acquisition system and a semantic parameter extension method for skinned multi-person linear model body type are proposed, which proves the practicality and scalability of the proposed method.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0178-2789
1432-2315
DOI:10.1007/s00371-022-02738-1