Automated Video Indexing for On-Demand Retrieval from Very Large Video Libraries

The Informedia Digital Video Library project is implementing full content search and retrieval from digital video, audio and text libraries. This is accomplished through the utilization of integrated speech, image and language understanding technologies for their automated creation and exploration. Image processing analyzes scenes, speech processing transcribes the audio signal, and natural language processing determines word relevance. Together, these generate a meaningful index into the video content. Segment breaks produced by image processing are examined along with the boundaries identified by the natural language processing of the transcript to partition the video library into sets of segments, or “video paragraphs”. Automating these techniques into a unified collaborative system enables us to include and search through vast amounts of video data in the library.