A Neurofuzzy Approach to Active Learning based Annotation Propagation for 3D Object Databases

Simple item page
dc.contributor.authorLazaridis, Michalisen_US
dc.contributor.authorDaras, Petrosen_US
dc.contributor.editorStavros Perantonis and Nikolaos Sapidis and Michela Spagnuolo and Daniel Thalmannen_US
dc.date.accessioned2013-10-21T18:15:19Z
dc.date.available2013-10-21T18:15:19Z
dc.date.issued2008en_US
dc.description.abstractMost existing Content-based Information Retrieval (CBIR) systems using semantic annotation, either annotate all the objects in a database (full annotation) or a manually selected subset (partial annotation) in order to increase the system's performance. As databases become larger, the manual effort needed for full annotation becomes unaffordable. In this paper, a fully automatic framework for partial annotation and annotation propagation, applied to 3D content, is presented. A part of the available 3D objects is automatically selected for manually annotation, based on their 'information content'. For the non-annotated objects, the annotation is automatically propagated using a neurofuzzy model, which is trained during the manual annotation process and takes into account the information hidden into the already annotated objects. Experimental results show that the proposed method is effective, fast and robust to outliers. The framework can be seen as another step towards bridging the semantic gap between low-level geometric characteristics (content) and intuitive semantics (context).en_US
dc.description.seriesinformationEurographics 2008 Workshop on 3D Object Retrievalen_US
dc.identifier.isbn978-3-905674-05-7en_US
dc.identifier.issn1997-0463en_US
dc.identifier.urihttps://doi.org/10.2312/3DOR/3DOR08/049-056en_US
dc.publisherThe Eurographics Associationen_US
dc.subjectCategories and Subject Descriptors (according to ACM CCS): I.5.2 [Pattern Recognition]: Design Methodologyen_US
dc.titleA Neurofuzzy Approach to Active Learning based Annotation Propagation for 3D Object Databasesen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
049-056.pdf
Size:
300.56 KB
Format:
Adobe Portable Document Format
Download
Collections
3DOR 08