Abstract: As the problem of the unsatisfied clothing segmentation accuracy, a novel segmentation algorithm is proposed based on HOG（Histogram of Oriented Gradients） features and E-SVM（Exemplar Support Vector Machine） classifiers. The co-segmentation algorithm can be divided into three steps：Superpixel grouping, E-SVM classifier training, segmentation propagation, which uses an auxiliary dataset. Firstly, user input image is segmented in superpixel, which makes the images in auxiliary dataset into regions. Secondly, some regions are selected to a positive segmentation, which uses HOG information to train E-SVM classifiers. Finally, the cloth is segmented in user image by E-SVM classifiers and segmentation propagation. Experimental results show that the proposed algorithm can segment clothing images with high accuracy.

Key words: co-segmentation, Histogram of Oriented Gradients（HOG） feature, superpixel grouping, Exemplar Support Vector Machine（E-SVM） classifier, segmentation propagation

摘要： 针对目前服装图像分割准确率低的问题，提出一种基于HOG特征和E-SVM分类器的服装图像联合分割算法。该算法具体可分为三个迭代的步骤：超像素组合、E-SVM分类器训练、分割传播，并用到辅助数据集。将用户输入的图像结合辅助服装集进行超像素分割，并利用分割传播方法将超像素组合成多个区域。利用分割效果积极的区域的HOG信息训练E-SVM分类器。通过E-SVM分类器以及分割传播方法将输入的图像中的服装分割出来。实验结果表明，该方法能够高准确率地分割出服装图像。

关键词: 联合分割, 方向梯度直方图（HOG）特征, 超像素组合, 模范支持向量机（E-SVM）分类器, 分割传播