Article
Computer Science, Information Systems
Han Yan, Haijun Zhang, Jianyang Shi, Jianghong Ma, Xiaofei Xu
Summary: In this research, a texture and shape disentangled generative adversarial network (TSD-GAN) is proposed for intelligent design in fashion items. TSD-GAN learns how to disentangle the features of texture and shape in an unsupervised manner and generates mixed-style fashion items, which can assist designers in the design process.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2023)
Article
Chemistry, Analytical
Leiyu Zhang, Zhenxing Jiao, Yandong He, Peng Su
Summary: This article introduces a soft exoskeleton for the hip flexion called H-Suit, which aims to improve walking endurance, delay muscle fatigue, and reduce the activation level of hip flexors. Through experimental methods and an assistance strategy, the desired profile of auxiliary forces is obtained and shown to be consistent with the theoretical values. The H-Suit is found to delay muscle fatigue of the lower limbs under powered conditions.
Article
Materials Science, Textiles
Yaping Li, Jie Fan, Shengyu Zhang, Zhaopeng Xia, Liang Wang, Yong Liu
Summary: This study presents a trilayered woven fabric with directional water transport performance by combining a hierarchical microporous structure and asymmetric wettability. The fabric exhibits excellent directional water management property, which can be controlled by the fineness and wetting state of the weft yarn, as well as the groove structure on the surface. The reported method for designing such fabric provides insights into developing eco-friendly water management textiles with good durability for demanding situations.
FIBERS AND POLYMERS
(2023)
Article
Computer Science, Information Systems
Xiaoling Gu, Jun Yu, Yongkang Wong, Mohan S. Kankanhalli
Summary: The proposed end-to-end network architecture can automatically synthesize photo-realistic fashion product images based on input poses and text descriptions, fooling the discriminator with a generator that predicts realistic-looking fashion images and an attention network enhancing the translation process. The discriminator is divided into multi-scale sub-discriminators to improve image distinguishing performance.
IEEE TRANSACTIONS ON MULTIMEDIA
(2021)
Article
Green & Sustainable Science & Technology
ElSayed A. Elnashar, Liliana Indrie, Dorina Camelia Ilies, Zlatin Zlatev, Grigore Vasile Herman, Cristina Secan, Jan A. Wendt
Summary: Investigating and collecting data on Romanian folk textile designs, particularly ornaments, colors, and patterns, is important for preserving traditional knowledge and adapting it for modern use. This research aims to collect and protect knowledge about old patterns and weaving techniques in Romania, and utilize them in contemporary folk handicraft weaving and the clothing industry.
Article
Computer Science, Artificial Intelligence
Shuhui Jiang, Jun Li, Yun Fu
Summary: This article proposes FashionG and SC-FashionG frameworks for generating fashion style images using deep neural network algorithms, effectively achieving single-style and mix-and-match style clothing image generation.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Information Systems
A. Vijayaraj, P. T. Vasanth Raj, R. Jebakumar, P. Gururama Senthilvel, N. Kumar, R. Suresh Kumar, R. Dhanagopal
Summary: This research uses convolutional neural networks to predict different fashion styles with high accuracy. Deep learning has been applied in the field of apparel manufacturing for clothing recognition, search, and suggestion. The experimental results show that the choice of activation function, optimizer, and dropout rate affects the accuracy of the results.
WIRELESS COMMUNICATIONS & MOBILE COMPUTING
(2022)
Article
Chemistry, Analytical
Leiyu Zhang, Xiang Gao, Ying Cui, Jianfeng Li, Ruidong Ge, Zhenxing Jiao, Feiran Zhang
Summary: A soft exoskeleton named A-Suit was developed to improve walking endurance and reduce metabolic consumption in the lower limbs by applying ergonomics design and biomechanical control methods. Performance experiments showed that wearing the A-Suit resulted in significant reductions in heart rate and metabolic consumption. This research provides valuable insights for human augmentation and medical assistance.
Article
Green & Sustainable Science & Technology
Mine Seckin, Ahmet Cagdas Seckin, Pinar Demircioglu, Ismail Bogrekci
Summary: This research proposes an approach to improve texture recognition and weaving parameter estimation in the textile industry. By using low-cost handheld microscopy and machine learning, significant achievements have been made, indicating promising applications.
Article
Computer Science, Artificial Intelligence
Caner Balim, Kemal Ozkan
Summary: Fashion image understanding is a popular research field, and this paper proposes a method of diagnosing outfit compatibility through clothing images. The proposed system not only predicts compatibility, but also diagnoses incompatible clothing items in outfits. A new dataset named ModAI was created, which includes clothing images and compatibility comments from different users. Image captioning techniques were used to generate compatibility suggestion texts from clothing images, and the model achieves a 0.62 BLEU-4 score. Experiments show that image captioning techniques can also be used to diagnose outfit compatibility.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Han Yan, Haijun Zhang, Linlin Liu, Dongliang Zhou, Xiaofei Xu, Zhao Zhang, Shuicheng Yan
Summary: The article introduces an AI-based framework for fashion design using generative adversarial networks to enhance designers' efficiency. The framework includes a sketch-generation module based on latent space and a rendering-generation module to learn the mapping between textures and sketches. Experimental results demonstrate the effectiveness of the proposed method in synthesizing semantic-aware textures on sketches.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Materials Science, Textiles
Xiaoxi Zhou, Yue Xu, Tui Chen
Summary: This study identified the interrelationship between users' perception image, suit design features, and users' preference by establishing PLS models, and proposed optimization schemes considering users' perception image and preference. The results indicate that the optimized designs significantly outperform other schemes.
INTERNATIONAL JOURNAL OF CLOTHING SCIENCE AND TECHNOLOGY
(2022)
Article
Computer Science, Information Systems
Fu-Hsien Huang, Hsin-Min Lu, Yao-Wen Hsu
Summary: Researchers are increasingly interested in using street photos to understand fashion trends, but face technical challenges such as noisy labels and the use of a multi-task learning framework. To address these challenges and understand Asian fashion trends, a new street fashion dataset called RichWear was created, and FARNet was proposed for fashion recognition.
Article
Computer Science, Information Systems
Xin Liu, Yongbin Sun, Ziwei Liu, Dahua Lin
Summary: This study introduces a novel fashion collocation framework, Neural Graph Filtering, which models a flexible set of fashion items using a graph neural network. The proposed approach significantly outperforms existing methods in terms of the AUC metric and is preferred by a majority of users for its diverse style recommendations.
IEEE TRANSACTIONS ON MULTIMEDIA
(2021)
Article
Materials Science, Textiles
Hiroyuki Kanai, Kentaro Ogawa, Tetsu Sasagawa, Kiyohiro Shibata
Summary: This study successfully produced worsted fabric with mechanical stretch property and verified the effect of this property on the comfort of men's suit pants.
TEXTILE RESEARCH JOURNAL
(2021)