Pattern Mining in Large-Scale Image and Video Sources
Detection and recognition of semantic events has been a major research challenge for multimedia indexing. An emerging direction in this field has been unsupervised discovery (mining) of patterns in spatial-temporal multimedia data. Patterns are recurrent, predictable occurrences of one or more entities that satisfy statistical, associative, or relational conditions. Patterns at the feature level may signify the occurrence of primitive events (e.g., recurrent passing of pedestrians). At the higher level, patterns may represent cross-event relations; e.g., recurrent news stories across multiple broadcast channels or repetitive play-break alternations in sports. Patterns in an annotated image collection may indicate collocations of related semantic concepts and perceptual clusters.