We present an interface for interactive video browsing where users visually skim storyboard representations of the files in search for known items (known-item search tasks) and textually described subjects, objects, or events (ad-hoc search tasks). Individual segments of the video are represented as a color-sorted storyboard that can be addressed via a color-index. Our storyboard representation is optimized for quick visual inspections considering results from our ongoing research. In addition, a concept based-search is used to filter out parts of the storyboard containing the related concept(s), thus complementing the human-based visual inspection with a semantic, content-based annotation.
CITATION STYLE
Hürst, W., Ching, A. I. V., Schoeffmann, K., & Primus, M. J. (2017). Storyboard-based video browsing using color and concept indices. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10133 LNCS, pp. 480–485). Springer Verlag. https://doi.org/10.1007/978-3-319-51814-5_45
Mendeley helps you to discover research relevant for your work.