One of the bottlenecks of current recognition (and graph matching) systems is their assumption of one-to-one feature (node) correspondence. This assumption breaks down in the gener...
Ali Shokoufandeh, Yakov Keselman, M. Fatih Demirci...
Abstract. We develop an object detection method combining top-down recognition with bottom-up image segmentation. There are two main steps in this method: a hypothesis generation s...
We introduce a new method for the recognition of partially occluded objects represented only by their contours. Object description, which stems from the inflection point detectio...
This paper presents a new shape registration algorithm that establishes "meaningful correspondence" between objects, in that it preserves the local shape correspondence ...
Yun Zhu, Xenophon Papademetris, Albert J. Sinusas,...