Stereoscopic Video Object Segmentation Based on SVM and Mean-Shift

Stereoscopic video object segmentation plays an important role in stereo vision analysis and application. In this paper, we proposed the combination support vector machine (SVM) and mean shift clustering algorithm to achieve classification and segmentation of stereoscopic video. We firstly divided the left image of stereoscopic video into the two classes of target and background and perform training and distinguishing by SVM. Then mean shift algorithm was exploited to achieve segmentation of disparity map. Finally, accurate objects were segmented according to fusion processing of SVM and disparity segmentation of mean shift. The experimental results show that the proposed method can segment the semantically meaningful objects successfully.