Abstract:We introduce an efficient and practical integrated system for human body model personalization with articulation. Starting with a 3D personalized model of the individual in a standard pose, the model is updated to accommodate for changes in articulations captured in a video clip with Ndocumentclass[12pt]{minimal}usepackage{amsmath}usepackage{wasysym}usepackage{amsfonts}usepackage{amssymb}usepackage{amsbsy}usepackage{mathrsfs}usepackage{upgreek}setlength{oddsidemargin}{-69pt}egin{document}$$N$$end{document} frames. The personalized model is segmented into different parts using anthropometric control points on the silhouette boundary of the frontal projection of the 3D model. These are endpoints of segments in 2D corresponding to projections of regions of independently moving parts in 3D. These points can either be manually selected or predicted using a pre-trained convolutional neural network (NN) point model. Model evolution consists of finding a set of 3D transformations that are independently applied to parts on 3D model so that the projections of the 3D model ‘match’ those observed in the video sequence at corresponding frames. This is done by minimizing the error between the frontally projected body region points and the target region points in the image for each independent moving part. The average vertex error of our articulation recovery method yields sub-resolution recovery errors ((about 4.77 mm compared to 17.82 mm—the resolution cell of the body model). This is quite an improvement over the SMPL NN approach using the same error metric that yields 10 times the resolution cell. The virtually reconstructed articulated 3D model is fitted with a 3D garment model for the creation of virtual fitting room that allows an individual to virtually access how well the garment fits.

Design of Robotic Mannequin Formed by Flexible Belt Net

The Design of a Deformable Robotic Mannequin

The Design of a Form-Changing Female Fitting Robot

Structural Design And Optimization Of A Panel-Based Fitting Robot

Computational Design of Shape-Changing Robotic Mannequin Based on 3D Human Models

Soft Robotic Mannequin: Design and Algorithm for Deformation Control

Personalized 3D Mannequin Reconstruction Based on 3D Scanning

Shape-Changing Robotic Mannequin Shoulder with Bio-Inspired Layered Structure

Design of a Deformable Lower Body Robot for Garment E-Commerce

Flexible Robotic Spine Actuated by Shape Memory Alloy

Generative Design of a Calf Structure for a Humanoid Robot Based on Gait Simulation

Design and Simulation Experiment of Rigid-Flexible Soft Humanoid Finger

Measuring and modeling human bodies with a novel relocatable mechatronic sensor-net

Controllable Stiffness Origami “skeletons” for Lightweight and Multifunctional Artificial Muscles

Advanced Artificial Muscle for Flexible Material-Based Reconfigurable Soft Robots.

Study on Parameterized Mannequin Modeling in 3-D Garment Simulation

Fitting 3D garment models onto individual human models

DeformerNet: Learning Bimanual Manipulation of 3D Deformable Objects

Virtual reconstruction of 3D articulated human shapes applied to garment try-on in a virtual fitting room

Research on 3D Virtual Design of Mannequin Specially for Model Group

Design of a multi-sensor information acquisition system for mannequin reconstruction and human body size measurement under clothes