Garment Image Structural Analysis Technology Research And Application

In recent years,computer technology is widely used in the clothing industry,such as garment CAD,modeling,sculpt and virtual try on. They all become the research hotspot. Obviously computer technology has pervaded every aspect of clothing industry. With the improvement in living standards,people's requirements have also increasingly higher. They don't require only for warm and simple prettiness,but pay more attention on individuation. This subject is from the fund based on the nature,the project's goals include morph or re-texture garments functional units for the implementation of personalize dress design .The main target of this subject is morph or re-texture garments components in two-dimensional images.This article is research in the following areas garments image segment technique,structural analysis of garment and re-texturing of garment image.(1) Garments image segment technique. this paper describes several garments image segment technique.We analyses their advantages and disadvantages, and then put forward using an interactive image segment technique to garments image segent. A two-stage method combing the region-level and pixel-level processing is proposed to segment the foreground object with the help of user's interaction. At first,watershed algorithm based on color feature is chosen to cluster the image into different regions. With the hint of user's interactive, a weighted graph is developed to minimize the Gibbs energy function to assign the binary labels to all regions by graph cut algorithm. Finally,a pixel-level graph around the foreground boundary band is developed to obtain the accurate foreground edge by pixel based graph cut algorithm.Experimental results demonstrate our method's efficiency and effectiveness. Experimental evidence this algorithm only need a little interactions and processing speedly and solves the problem and methods based on model's final result are used as to limitation by early detection and treatment.(2) Structural analysis of garment. Garment parts are selected by structural analysis of garment, and concrete method is as follow : first we build the dress sketch template, by matching the template with the clothing in the image,then we can bring the internal structure information of the template into cloth in the image by Complex Barycentric Coordinates .By the internal structure information we can segment the garment parts.(3) Re-texturing of garment image. Experimental demonstrate our method achieves in garments and garment's parts re-texturing , this result shows the signification of structural analysis of garment. This thesis introduces a novel image retexturing approach that preserves the original shading effects without knowing the underlying surface and lighting conditions. In this approach, a Poisson equation-based algorithm is developed to simulate the texture distortion on the projected interest region of the underlying surface, while shading effect of the original image is preserved through YCbCr color space translation and transfer.