Virtual Fitting System Of 3D Segmentation Technique Based On Kinect Device

With the development of Internet and electronic commerce industry,a new way of trading-online shopping has become a mainstream shopping way,making virtual fitting become a hot topic in recent years.In the field of image processing and computer vision,image segmentation has been one of the most difficult tasks.As a key technology,it directly determines the effectiveness of the image processing.With the development of the innovation and the development of the hardware technology,the Microsoft Corp which had launched the Kinect device greatly reduces the cost of three-dimensional image research,and thus greatly promotes the research in the field of three-dimensional image.In this paper,we summarize the existing 3D segmentation techniques,and propose an improved algorithm based on the SLIC algorithm.The K-means clustering algorithm is based on the points cloud data obtained by Kinect.The proposed method has a significant improvement in the boundary recall and the under-segmentation error.In order to be able to match the subsequent virtual fitting system and the clothing size completely,we extract human body feature measurement via the feature point and feature surface feature based on the previous work.Through the comparison of the experimental data,it is verified that our method is a good improvement.