作者: Hansung Kim , Dong Bo Min , Kwanghoon Sohn
DOI: 10.1007/11581772_34
关键词: Computer vision 、 Image processing 、 Artificial intelligence 、 Process (computing) 、 Computer science 、 Subtraction 、 Initialization 、 Background subtraction 、 Matching (graph theory) 、 Pattern matching 、 Segmentation
摘要: We propose a fast disparity estimation algorithm using background registration and object segmentation for stereo sequences from fixed cameras. Dense information is calculated in an initialization step so that only disparities of moving regions are updated the main process. real-time technique subtraction inter-frame differences, hierarchical region-dividing shape-adaptive matching windows. Experimental results show proposed provides accurate vector fields with average processing speed 15 frames/sec 320x240 on common PC.