A cluster-based statistical model for object detection

T.D. Rikert, M.J. Jones, P. Viola
1999 Proceedings of the Seventh IEEE International Conference on Computer Vision  
This paper presents an approach to object detection which is based on recent work in statistical models for texture synthesis and recognition 7, 4, 23, 17 . Our method follows the texture r ecognition work of De Bonet and Viola 4 . We use feature vectors which capture the joint occurrence o f l o cal features at multiple resolutions. The distribution of feature v e ctors for a set of training images of an object class is estimated by clustering the data and then forming a mixture o f gaussian
more » ... del. The mixture m o del is further re ned by determining which clusters are the most discriminative for the class and retaining only those clusters. After the model is learned, test images are classi ed b y computing the likelihood of their feature vectors with respect to the model. We present promising results in applying our technique to face detection and car detection.
doi:10.1109/iccv.1999.790386 dblp:conf/iccv/RikertJV99 fatcat:yowwrefehvfcreuym3n5bk7oem