TY - CONF AU - Antonio Hernandez AU - Nadezhda Zlateva AU - Alexander Marinov AU - Miguel Reyes AU - Petia Radeva AU - Dimo Dimov AU - Sergio Escalera A2 - CVPR PY - 2012// TI - Graph Cuts Optimization for Multi-Limb Human Segmentation in Depth Maps BT - 25th IEEE Conference on Computer Vision and Pattern Recognition SP - 726 EP - 732 PB - IEEE Xplore N2 - We present a generic framework for object segmentation using depth maps based on Random Forest and Graph-cuts theory, and apply it to the segmentation of human limbs in depth maps. First, from a set of random depth features, Random Forest is used to infer a set of label probabilities for each data sample. This vector of probabilities is used as unary term in α-β swap Graph-cuts algorithm. Moreover, depth of spatio-temporal neighboring data points are used as boundary potentials. Results on a new multi-label human depth data set show high performance in terms of segmentation overlapping of the novel methodology compared to classical approaches. SN - 1063-6919 SN - 978-1-4673-1226-4 L1 - http://refbase.cvc.uab.es/files/HZM2012b.pdf UR - http://dx.doi.org/10.1109/CVPR.2012.6247742 N1 - MILAB;HuPBA ID - Antonio Hernandez2012 ER -