Abstract:
We present a novel feature representation for categorical object detection. Unlike previous approaches that have concentrated on generic interest-point detectors, we cons...Show MoreMetadata
Abstract:
We present a novel feature representation for categorical object detection. Unlike previous approaches that have concentrated on generic interest-point detectors, we construct object-specific features directly from the training images. Our feature is represented by a collection of Flexible Edge Arrangement Templates (FEATs). We propose a two-stage semi-supervised learning approach to feature selection. A subset of frequent templates are first selected from a large template pool. In the second stage, we formulate feature selection as a regression problem and use LASSO method to find the most discriminative templates from the preselected ones. FEATs adaptively capture the image structure and naturally accommodate local shape variations. We show that this feature can be complemented by the traditional holistic patch method, thus achieving both efficiency and accuracy. We evaluate our method on three well-known car datasets, showing performance competitive with existing methods.
Published in: 2008 IEEE Workshop on Applications of Computer Vision
Date of Conference: 07-09 January 2008
Date Added to IEEE Xplore: 17 June 2008
ISBN Information:
Print ISSN: 1550-5790