This paper describes an efficient method to obtain 3D information by using spatio-temporal analysis of omni images for outdoor navigation and map-making in the intelligent transpo...
Inspired by the underlying relationship between classification capability and the mutual information, in this paper, we first establish a quantitative model to describe the inform...
Determining the correspondence of image patches is one of the most important problems in Computer Vision. When the intensity space is variant due to several factors such as the ca...
Kernel methods have been widely studied in the field of pattern recognition. These methods implicitly map, "the kernel trick," the data into a space which is more approp...
Recovering the three-dimensional (3D) object shape lies as an unresolved and active research topic on the crosssection of computer vision, photogrammetry and bioinformatics. Altho...