Pii: S0031-3203(01)00163-7
نویسندگان
چکیده
Search by object model — 3nding an object inside a target image — is a desirable and yet di4cult mechanism for querying multi-media data. An added di4culty is that objects can be photographed under di8erent lighting conditions. While human vision has color constancy, an invariant processing, presumably, here we seek only covariant processing and look to recover such lighting change. Making use of feature-consistent locales in an image we develop a scene partition by localization, rather than by image segmentation. A diagonal model for illumination change and a voting scheme in chromaticity space provide a candidate set of lighting change coe4cients for covariant image transformation. For each pair of coe4cients, Elastic Correlation, a form of correlation of locale colors, is performed along with a least squares minimization for pose estimation. Since the rotation, scale and translation parameters are thus estimated, we can apply an e4cient process of texture support and shape veri3cation. Tests on an image and video database of about 1500 images show an average recall and precision of over 70%. ? 2002 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved.
منابع مشابه
Illumination color covariant locale-based visual object retrieval
Search by Object Model — finding an object inside a target image — is a desirable and yet difficult mechanism for querying multimedia data. An added difficulty is that objects can be photographed under different lighting conditions. While human vision has color constancy, an invariant processing, presumably, here we seek only covariant processing and look to recover such lighting change. Making...
متن کاملPii: S0031-3203(01)00129-7
This paper presents a survey on o!-line Cursive Word Recognition. The approaches to the problem are described in detail. Each step of the process leading from raw data to the /nal result is analyzed. This survey is divided into two parts, the /rst one dealing with the general aspects of Cursive Word Recognition, the second one focusing on the applications presented in the literature. ? 2002 Pat...
متن کاملGaussian mixture parameter estimation with known means and unknown class-dependent variances
متن کامل
Visual Segment Tree Creation for MPEG-7 Description Schemes
This paper deals with the creation of visual segment trees involved in MPEG-7 Description Schemes. After a brief overview of MPEG7 description schemes in general and of the Segment Description Scheme in particular, tools allowing the creation of segment trees are discussed. It is proposed to create a Binary Partition Tree in a first step and to restructure the tree in a second step. Several exa...
متن کامل