Image and Video Retrieval: Third International Conference, CIVR 2004, Dublin, Ireland, July 21-23, 2004, Proceedings [Paperback]

Keynote Speaker Abstracts.- Pattern Mining in Large-Scale Image and Video Sources.- Computer Vision in the Movies: From the Lab to the Big Screen.- Image and Video Retrieval Using New Capture and Display Devices.- Models of Interaction with Video Information.- Image Annotation and User Searching.- User Strategies in Video Retrieval: A Case Study.- Video Content Foraging.- Using Maximum Entropy for Automatic Image Annotation.- Everything Gets Better All the Time, Apart from the Amount of Data.- Image and Video Retrieval Algorithms (I).- An Inference Network Approach to Image Retrieval.- Small Sample Size Performance of Evolutionary Algorithms for Adaptive Image Retrieval.- Co-retrieval: A Boosted Reranking Approach for Video Retrieval.- Poster Session (I).- HMM Model Selection Issues for Soccer Video.- Tennis Video Analysis Based on Transformed Motion Vectors.- Semantic Event Detection in Sports Through Motion Understanding.- Structuring Soccer Video Based on Audio Classification and Segmentation Using Hidden Markov Model.- EDU: A Model of Video Summarization.- A News Video Mining Method Based on Statistical Analysis and Visualization.- Topic Threading for Structuring a Large-Scale News Video Archive.- Whats News, Whats Not? Associating News Videos with Words.- Visual Clustering of Trademarks Using a Component-Based Matching Framework.- Assessing Scene Structuring in Consumer Videos.- A Visual Model Approach for Parsing Colonoscopy Videos.- Video Summarization and Retrieval System Using Face Recognition and MPEG-7 Descriptors.- Automatic Generation of Personalized Digest Based on Context Flow and Distinctive Events.- Content-Based Image Retrieval and Characterization on Specific Web Collections.- Exploiting Problem Domain Knowledge for Accurate Building ImageClassification.- Natural Scene Retrieval Based on a Semantic Modeling Step.- Unsupervised Text Segmentation Using Color and Wavelet Features.- Universal and Personalized Access to Content via J2ME Terminals in l2