Search

Human body precise segmentation is difficult because of inter-occlusion when there are multiple human bodies in video. A coarse-to-fine segmentation method is proposed. In coarse segmentation, human shape models are used to get human’s position and coarse region. The human models with variant scale and posture are constructed with head, torso, and legs. For each human body, its corresponding human shape model is obtained by model matching, and by which human position is obtained roughly. Human precise contour is obtained in fine segmentation by curve evolution with initial contour obtained from coarse segmentation. Experiment results show that the proposed method could segment human object precisely. DOI : http://dx.doi.org/10.11591/telkomnika.v12i5.3539

A method based on multi-feature and multi-stage verification is proposed in this paper. The direction and texture features obtained by Gabor filter are used for the first-stage verification. Then the sham dynamic features representing the writing force are used for the second-stage verification. Considering the number of samples for training is limited, the support vector machine (SVM) is selected as the classifier. The experimental results show that this method can not only increase the accuracy, but also satisfy the requirement of speed in real-time system as well.