Existing works about fashion outfit compatibility focus on predicting the overall of a set items with their information from different modalities. However, there are few explore how to explain prediction, which limits persuasiveness and effectiveness model. In this work, we propose an approach not only predict but also diagnose compatibility. We introduce end-to-end framework for goal, features for: (1) The is learned all type-specified pairwise similarities between items, backpropagation...

Using compound microscopy is one of the major options for identification cashmere/wool. To interpret human perception via machine vision, microscopic images captured by a charge-coupled device camera were transferred into projection curves. Three different deciphering methods, recurrence quantification analysis, direct geometrical description, and discrete wavelet transform employed to reveal embedded numerical features. The extracted parameters used screen supervised classification...

Purpose This paper presents methods and algorithms to automatically segment measure the human body. Design/methodology/approach In segmentation procedure, two different are designed find crotch point for situation of non‐contacted thigh contacted thigh, respectively. Three methods: minimum distance algorithm, inclination angle directional neighbor identification algorithm introduced search branching points or triangle. body measurement a pre‐sorted circling method is circumference...

It is an important topic to characterize the cross-sectional shape of profiled fiber for quality control textiles and performance design. In this paper, we propos e a new measure factor, which can be used cross-sections fibers. Different from existing indices factor are based on ratio radii circumscribed inscribed circles, proposed uses information residing in boundary cross-sections. defined as coefficient variation area sequence plane at each angular interval. The capacity validated by...

The garment dressing simulation is a process which transforms multiple two-dimensional patterns into three-dimensional configuration that follows the surface of human body and enables visualization draping effect garment. dimensional stability pattern an important constraint needs to be imposed ensure created has size complying with original design pattern. In this paper, we used mass-spring system simulate physical properties fabric adaptive sewing forces wrap around model form virtual...

Purpose Currently, a common method of reconstructing mannequin is based on the body measurements or features, which only preserve size lacking accurate geometric shape information. However, same human measurement does not equal to shape. This may result in an unfit garment for target body. The purpose this paper propose novel scanning-based pipeline reconstruct personalized mannequin, preserves both and Design/methodology/approach authors first capture subject via 3D scanning, statistical...

Due to the similarities between cashmere and wool, automatic identification of these two animal fibers continues be a huge challenge in textile society. In this paper, for micrographs bag-of-words spatial pyramid matching are used. Each fiber image was regarded as collection feature vectors our logic. The vectors, extracted from original dataset, were fed into support vector machine supervised classification. codebook size resolution level completely investigated. experimental results...

In this paper, a system with six depth cameras was built to scan both feet simultaneously. An improved calibration method based on T-shaped checkerboard used calculate the extrinsic parameters of cameras. virtual checkerboards were introduced further fine-tune accuracy iterative closest point algorithm. Based proposed foot scanner, complete procedure measure automatically by locating anatomical landmarks without manual intervention. Various experiments presented validate performance scanner...

Pattern recognition and feature extraction methods are applied to identify cashmere wool fibers, which two kinds of very similar animal fibers. In this article, we proposed a new identification method based on Speed Up Robust Features fiber images. The images fibers obtained by scanning electron microscopy. extracted, each image is regarded as collection vectors in our logic. fed into support vector machine for supervised learning. findings from microscope indicate that effective; the rate...

Image-based garment transfer systems aim to swap the desired clothes from a model arbitrary users. However, existing works cannot provide capacity for users try on various fashion articles according their wishes, i.e., can decide which article (e.g., tops, pants or both) be swapped. In this paper, we propose an Inpainting-based Virtual Try-On Network (I-VTON) allows user image in selective manner. To realize selectivity, reshape virtual try-on as task of inpainting. Firstly, texture and are...

In this paper, a new method was proposed to establish the relationship between three-dimensional (3D) foot shapes and their two-dimensional (2D) silhouettes, through which complete 3D shape can be predicted by simply inputting its two 2D silhouettes. scans of 80 participants were randomly selected as training set, those another 20 used testing set. Elliptical Fourier analysis (EFA) principle component (PCA) adopted parameterize shapes. A linear regressive model then developed predict with...

In apparel manufacturing, more than 50% of the total cost is consumed by fabric in comparison to rest materials. The efficient use may result reduction considerable expenditure and efforts. Therefore, purpose this study propose an image projected virtual marker (IPVM) onto surface layers/plies improve utilization visual quality control. 20 garment production orders are scheduled for experimental work investigate fabrics with imperfections. statistical analysis indicates a 3.5%–4% improvement...

Purpose The automatic body measurement is the key of tailoring, mass customization and fit/ease evaluation. major challenges include finding landmarks extracting sizes accurately. purpose this paper to propose a new method based on loop structure. Design/methodology/approach scanned human model sliced equally layers consist various shapes loops. semantic feature analysis has been regarded as problem points interest (POI) (LOI) according types connections. Methods for determining basic have...

In this article, we presented a new automatic three-dimensional-scanned garment fitting method for A-Pose-scanned human models. Both the and body were decomposed based on feature lines defined by various landmarks. The patches of three-dimensional automatically positioned around model setting up correspondence via matching. Virtual sewing was engaged to obtain final results virtual dressing. penetration between cloth solved geometrical constrained Laplacian-based deformation. experimental...

The present study deals with the extraction of eco-friendly natural dye from pomegranate peel through pre-mordanting method and its application on Tencel fabric. investigation was made for evaluation best processes (solvent, ultrasonic aqueous) use three mordants (stannous chloride, ferrous sulfate alum) to compare effect different color shades fastness properties. experimental result indicated that solvent delivered better effects fabric as compared aqueous (solvent[Formula: see...

Dressing up is often conducted in a layered manner. People would like various combinations of garments to be tried reach maximum satisfaction. To simulate this behavior, we propose here method for dressing an avatar with multi-layered virtual garments. The intersection among was located via Axis Aligned Bounding Box (AABB) accelerated detection, and the penetrations garment layers were rectified by forcing intersected area one party move towards correct side along shape other party. results...

The online purchase of garments is increasing and a method to enable accurate immediate garment sizing could improve the customer’s experience. A successful shopping system should provide capacity dressing given onto various posed human models with fit/ease information. In this paper, we present “copy” pose source model target via skeleton-matching algorithm. skeleton generated automatically according anthropometric features. difference compensated by an affine transformation applied skin...

In this paper, 3D fabric drape was reconstructed by using video recorded from a smartphone. Elliptical Fourier Analysis (EFA) and Principle Component (PCA) were used to parameterize the reveal shape parameters. A cluster analysis of various drapes implemented verify proposed method. Experiment results demonstrated that can be parameterized with mean error 0.52 mm when harmonic number EFA equals 25. The result indicated new features detected our method useful classify different drapes, which...

The 3D virtual mannequin has been widely used in apparel industry, and its importance is also increasing. This work develops a new reconstruction system based on optimization. All the mannequins reconstructed by proposed approach share identical topology, that is, there point-to-point correspondence among mannequins, which will significantly facilitate much subsequent processing fashion design, made-to-measure, try-on. inputs to contain template human body, raw scan (represented mesh), very...

Purpose The paper aims to transfer the item image of a given clothing product corresponding area user image. Existing classical methods suffer from unconstrained deformation and occlusion caused by hair or poses, which leads loss details in try-on results. In this paper, authors present details-oriented virtual network (DO-VTON), allows synthesizing high-fidelity images with preserved characteristics target clothing. Design/methodology/approach proposed consists three modules. fashion...