A translation approach to portable ontology specifications
Knowledge Acquisition - Special issue: Current issues in knowledge modeling
FotoFile: a consumer multimedia organization and retrieval system
Proceedings of the SIGCHI conference on Human Factors in Computing Systems
Learning primitive and scene semantics of images for classification and retrieval
MULTIMEDIA '99 Proceedings of the seventh ACM international conference on Multimedia (Part 2)
MyPhotos: a system for home photo management and processing
Proceedings of the tenth ACM international conference on Multimedia
Semantic Home Photo Categorization
IEEE Transactions on Circuits and Systems for Video Technology
Hi-index | 0.00 |
This paper presents a prototype system that helps users manage and search collections of their digital photos. Existing systems and methods suffer from some shortcomings. For example, the visual similarity based search has poor precision and lacks for semantic information, while the text-based retrieval is only lexically motivated and those lexicons are not organized well. In one word, systems that really support intelligent retrieval are rare. In order to overcome these shortcomings, an ontology based approach is proposed for photo management, which is conceptual motivated. The approach is implemented in the OntoAlbumsystem. When a photo is imported into the database, the system automatically creates an instance of it. Then the user annotates the photo with concepts of ontology. Finally, by clicking on an instance of concepts in the ontology, the user can find all the photos related to the instance through ontology inference. Ontology provides a formal, explicit and conceptual annotation as well as an effective inference query. It makes intelligent retrieval possible in the photo management system. Experiments on the OntoAlbumsystem show that the proposed approach is very effective and promising.