In this paper, we combine discrete and continuous image models with information-theoretic-based criteria for unsupervised hierarchical image-set clustering. The continuous image modeling is based on mixture of Gaussian densities. The unsupervised image-set clustering is based on a generalized version of a recently introduced information-theoretic principle, the information bottleneck principle. Images are clustered such that the mutual information between the clusters and the image content is maximally preserved. Experimental results demonstrate the performance of the proposed framework for image clustering on a large image set. Information theoretic tools are used to evaluate cluster quality. Particular emphasis is placed on the application of the clustering for efficient image search and retrieval.

Abstract-The technical work on the first amendment of the H.264/MPEG4-AVC video coding standard has recently been completed. In these so-called Fidelity Range Extensions (FRExt) a set of new coding tools is specified which is primarily targeted at providing significant improvements in coding efficiency for higher-fidelity video material. This paper presents an overview of the corresponding methods, briefly discusses some important aspects regarding profiles and applications, and finally provides experimental results for a performance comparison with existing coding technology.

BackgroundPrimary headaches and Learning difficulties are both common in the pediatric population. The goal of our study was to assess the prevalence of learning disabilities and attention deficit disorder in children and adolescents with migraine and tension type headaches.MethodsRetrospective review of medical records of children and adolescents who presented with headache to the outpatient pediatric neurology clinics of Bnai-Zion Medical Center and Meyer Children’s Hospital, Haifa, during the years 2009–2010. Demographics, Headache type, attention deficit disorder (ADHD), learning disabilities and academic achievements were assessed.Results243 patients met the inclusion criteria and were assessed: 135 (55.6%) females and 108 (44.4%) males. 44% were diagnosed with migraine (35.8% of the males, 64.2% of the females, p = 0.04), 47.7% were diagnosed with tension type headache (50.4% of the males, 49.6% of the females). Among patients presenting with headache for the first time, 24% were formerly diagnosed with learning disabilities and 28% were diagnosed with attention deficit disorder (ADHD). ADHD was more prevalent among patients with tension type headache when compared with patients with migraine (36.5% vs. 19.8%, p = 0.006). Poor to average school academic performance was more prevalent among children with tension type headache, whereas good to excellent academic performance was more prevalent among those with migraine.ConclusionsLearning disabilities and ADHD are more common in children and adolescents who are referred for neurological assessment due to primary headaches than is described in the general pediatric population. There is an association between headache diagnosis and school achievements.

Obesity and primary headaches in children are associated. Although obesity seems to be a risk factor for migraine more than for tension-type headache, it is associated with increased headache frequency and disability regardless of headache type.