In recent years, single image super-resolution (SISR) methods using deep convolution neural network (CNN) have achieved impressive results. Thanks to the powerful representation capabilities of the deep networks, numerous previous ways can learn the complex non-linear mapping between low-resolution (LR) image patches and their high-resolution (HR) versions. However, excessive convolutions will limit the application of super-resolution technology in low computing power devices. Besides, super-resolution of any arbitrary scale factor is a critical issue in practical applications, which has not been well solved in the previous approaches. To address these issues, we propose a lightweight information multi-distillation network (IMDN) by constructing the cascaded information multidistillation blocks (IMDB), which contains distillation and selective fusion parts. Specifically, the distillation module extracts hierarchical features step-by-step, and fusion module aggregates them according to the importance of candidate features, which is evaluated by the proposed contrast-aware channel attention mechanism. To process real images with any sizes, we develop an adaptive cropping strategy (ACS) to super-resolve block-wise image patches using the same well-trained model. Extensive experiments suggest that the proposed method performs favorably against the state-of-the-art SR algorithms in term of visual quality, memory footprint, and inference time. Code is available at https://github.com/Zheng222/IMDN.

Purpose -The paper aims to provide an overview of the area of digital pattern developing for customized apparel. Design/methodology/approach -The paper outlines several methods of digital pattern developing for customized apparel, and discusses the principles, characters and applications. Digital pattern developing process has two paths. One path develops apparel according to traditional 2D pattern-making technology. There are three methods: parametric design, traditional grading technique, and pattern generating based on artificial intelligence (AI). Another path develops pattern through surface flattening directly from individual 3D apparel model. Findings -For parametric method, it can improve greatly the efficiency of pattern design or pattern alteration. However, the development and application of parametric Computer-Aided-Design (CAD) systems in apparel industry are difficult, because apparel pattern has fewer laws in graphical structure. For grading technique, it is the most practical method because of its simple theory, with which pattern masters are familiar. But these methods require users with higher experience. Creating expert pattern system based on AI can reduce the experience requirements. Meanwhile, a great deal of experiments should be conducted for each garment with different style to create their knowledge databases. For 3D CAD technology, two methods of surface flattening have been outlined, namely geometry flattening and physical flattening. But many improvements should be done if the 3D CAD systems are applied in apparel mass customization. Originality/value -The paper provides information of value to the future research on developing a practical made-to-measure apparel pattern system.

Purpose -In order to mass-customize clothes, it is essential to create prototype pattern according to individual body shape. The purpose of this paper is to present a new method to generate prototype pattern based on individual three-dimensional (3D) virtual dummy for further study on apparel customization. Design/methodology/approach -The symmetrized preprocessing and convex hull method are employed to create a dress-like virtual dummy based on 3D body scanning data. The corresponding structure lines of 2D prototype pattern are defined on the 3D dummy in advance and 3D dummy surface (only half) is cut into ten zones. Based on the characteristics of each surface, further subdivision was made in each zone to create 3D wireframe of garment prototype by calculating the intersection curves between the dummy surface and local planners. Via flattening geometrically 3D wireframe of each zone, final pattern of the prototype is got. Moreover, during the course of flattening of each zone, define constrained lines in advance so as to ensure the position and direction of each cutting pattern beforehand. Findings -The paper finds that 2D cutting patterns of the prototype have been constructed from the computerized 3D dummy. The length of major structure lines for both 3D model and 2D cutting pattern remain the same. The seven out of ten of cutting patterns have area error within^1 cm 2 compared to 3D surface. Only two cutting have relatively larger error but controlled within 3 cm 2 . Originality/value -The most outstanding property of the method developed is the possibility of geometrical transformation of 3D surface to 2D pattern through constructing 3D wireframe of the prototype garment, with no need to define physical-mechanical properties of fabric used. The newly created 2D cutting patterns have the coincident construction and shape with conventional prototype and are of outstanding quality and preciseness.

No abstract

No abstract