Abstract
This paper presents a variant of our generative probabilistic multimedia retrieval model. Evaluation on the TRECVID 2003 collection shows the new variant, a document generation approach, is suitable for information needs with multiple examples. Moreover, in combination with textual information, the new variant outperforms the original one. © Springer-Verlag 2004.
Cite
CITATION STYLE
APA
Westerveld, T., & Arjen, P. (2004). Multimedia Retrieval Using Multiple Examples. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3115, 344–352. https://doi.org/10.1007/978-3-540-27814-6_42
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
Already have an account? Sign in
Sign up for free