ABSTRACT

This chapter addresses the problems in broad by describing three basic components: content structuring and organization, data cleaning, and summarization, to enable management of large digital media archival. Digital media such as images and videos tell a story by showing millions of pixels in different patterns and at different snapshots of time. With decades of research efforts in multimedia community, general solutions to the aforementioned problem include the extraction of indexable visual patterns and the generation of short summaries to facilitate efficient browsing of digital media content. The effective management of digital media archival becomes even challenging, with the proliferation of social media websites and the arrival of massive multimedia data in these sites. Digital video archival such as web and news videos often contain large sets of duplicate or near-duplicate data. To detect near-duplicates and eventually exclude them from further processing in search involves the analysis of visual content.