The major problem in building a good lipreading system is to extract effective visual features from enormous quantity of video sequences data. For appearance-based feature analysi...
Yun Fu, Xi Zhou, Ming Liu, Mark Hasegawa-Johnson, ...
This paper presents a new algorithm for the automatic recognition of object classes from images (categorization). Compact and yet discriminative appearance-based object class mode...
Abstract. We present an approach to determine the category and location of objects in images. It performs very fast categorization of each pixel in an image, a brute-force approach...
Top-down visual saliency facilities object localization by providing a discriminative representation of target objects and a probability map for reducing the search space. In this...
We present a novel categorical object detection scheme that uses only local contour-based features. A two-stage, partially supervised learning architecture is proposed: a rudiment...