en
Proceedings chapter
Open access
English

Interactive Video Retrieval based on Multimodal Dissimilarity Representation

Presented at Bonn (Germany)
Publication date2005
Abstract

We present an approach to learn user semantic queries from dissimilarity representations of video audio-visual content. When dealing with large corpora of videos documents, using a feature-based representation calls for the online computation of distances between all documents and the query. Hence, a dissimilarity representation may be preferred because its offline computation speeds up the retrieval process. We show how distances related to visual and audio video features can directly be used to learn complex concepts from a set of positive and negative examples provided by the user. Based on the idea of dissimilarity spaces, we derive a lowdimensional multimodal representation space where an on-line and real-time classification is performed to learn user queries. The classification consists in maximizing a non-linear Fisher criterion to separate positive from negative examples. The evaluation, performed on the complete annotated TRECVid corpus, shows that our technique enables us to improve the precision of retrieval results.

Keywords
  • multimodal dissimilarity representation
  • interactive video retrieval
Citation (ISO format)
BRUNO, Eric, MOENNE-LOCCOZ, Nicolas, MARCHAND-MAILLET, Stéphane. Interactive Video Retrieval based on Multimodal Dissimilarity Representation. In: Proceedings of the Machine Learning Techniques for Processing Multimedia Content- ICML workshop, MLMM′05. Bonn (Germany). [s.l.] : [s.n.], 2005.
Main files (1)
Proceedings chapter (Published version)
accessLevelPublic
Identifiers
  • PID : unige:47729
299views
73downloads

Technical informations

Creation2015/03/06 17:12:10
First validation2015/03/06 17:12:10
Update time2023/03/14 22:58:53
Status update2023/03/14 22:58:53
Last indexation2024/01/16 17:09:25
All rights reserved by Archive ouverte UNIGE and the University of GenevaunigeBlack