Purpose: Currently, a common method of reconstructing mannequin is based on the body measurements or body features, which only preserve the body size lacking of the accurate body geometric shape information. However, the same human body measurement does not equal to the same body shape. This may result in an unfit garment for the target human body. The purpose of this paper is to propose a novel scanning-based pipeline to reconstruct the personalized mannequin, which preserves both body size and body shape information. Design/methodology/approach: The authors first capture the body of a subject via 3D scanning, and a statistical body model is fit to the scanned data. This results in a skinned articulated model of the subject. The scanned body is then adjusted to be pose-symmetric via linear blending skinning. The mannequin part is then extracted. Finally, a slice-based method is proposed to generate a shape-symmetric 3D mannequin. Findings: A personalized 3D mannequin can be reconstructed from the scanned body. Compared to conventional methods, the method can preserve both the size and shape of the original scanned body. The reconstructed mannequin can be imported directly into the apparel CAD software. The proposed method provides a step for digitizing the apparel manufacturing. Originality/value: Compared to the conventional methods, the main advantage of the authors’ system is that the authors can preserve both size and geometry of the original scanned body. The main contributions of this paper are as follows: decompose the process of the mannequin reconstruction into pose symmetry and shape symmetry; propose a novel scanning-based pipeline to reconstruct a 3D personalized mannequin; and present a slice-based method for the symmetrization of the 3D mesh.

In the wake of the use of deep learning algorithms in medical image analysis, we compared performance of deep learning algorithms, namely the deep Boltzmann machine (DBM), convolutional encoder network (CEN) and patch-wise convolutional neural network (patch-CNN), with two conventional machine learning schemes: Support vector machine (SVM) and random forest (RF), for white matter hyperintensities (WMH) segmentation on brain MRI with mild or no vascular pathology. We also compared all these approaches with a method in the Lesion Segmentation Tool public toolbox named lesion growth algorithm (LGA). We used a dataset comprised of 60 MRI data from 20 subjects in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database, each scanned once every year during three consecutive years. Spatial agreement score, receiver operating characteristic and precision-recall performance curves, volume disagreement score, agreement with intra-/inter-observer reliability measurements and visual evaluation were used to find the best configuration of each learning algorithm for WMH segmentation. By using optimum threshold values for the probabilistic output from each algorithm to produce binary masks of WMH, we found that SVM and RF produced good results for medium to very large WMH burden but deep learning algorithms performed generally better than conventional ones in most evaluations.

We introduce a data-driven method to generate a large number of plausible, closely interacting 3D human pose-pairs, for a given motion category, e.g., wrestling or salsa dance. With much difficulty in acquiring close interactions using 3D sensors, our approach utilizes abundant existing video data which cover many human activities. Instead of treating the data generation problem as one of reconstruction, either through 3D acquisition or direct 2D-to-3D data lifting from video annotations, we present a solution based on Markov Chain Monte Carlo (MCMC) sampling. With a focus on efficient sampling over the space of close interactions, rather than pose spaces, we develop a novel representation called interaction coordinates (IC) to encode both poses and their interactions in an integrated manner. Plausibility of a 3D pose-pair is then defined based on the ICs and with respect to the annotated 2D pose-pairs from video. We show that our sampling-based approach is able to efficiently synthesize a large volume of plausible, closely interacting 3D pose-pairs which provide a good coverage of the input 2D pose-pairs.