Digital library access via image similarity search

Summary from only given. We present an image library with similarity-based searching and novel browsing structures and a news video library with content-based search and summarisation features. Our approach puts the traditional library search via textual meta-data upside down by allowing access to collections through browsing and search of images that may come as part of a multimedia document. We have developed two systems, which address the need for nontextual search facilities in multimedia collections. The first is an integrated system for building and providing access to a broad range of image collections such as engineering sketches, wild-life photographs or private photo collections. The second system called ANSES (Automatic News Summarization Extraction System) is explicitly targeting multimedia information pertaining to video broadcasts. ANSES captures television news each day along with the associated subtitles.