Semantic based image retrieval: a probabilistic approach
MULTIMEDIA '00 Proceedings of the eighth ACM international conference on Multimedia
Content-Based Image Retrieval at the End of the Early Years
IEEE Transactions on Pattern Analysis and Machine Intelligence
Coupled hidden Markov models for complex action recognition
CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
A Bayesian Approach to Unsupervised One-Shot Learning of Object Categories
ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
Video Data Mining: Semantic Indexing and Event Detection from the Association Perspective
IEEE Transactions on Knowledge and Data Engineering
Beyond pixels: Exploiting camera metadata for photo classification
Pattern Recognition
Object-based image retrieval using the statistical structure of images
CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
A probabilistic framework for semantic video indexing, filtering,and retrieval
IEEE Transactions on Multimedia
IEEE Transactions on Multimedia
Measuring Concept Similarities in Multimedia Ontologies: Analysis and Evaluations
IEEE Transactions on Multimedia
Mining Multilevel Image Semantics via Hierarchical Classification
IEEE Transactions on Multimedia
Image classification for content-based indexing
IEEE Transactions on Image Processing
CBSA: content-based soft annotation for multimodal image retrieval using Bayes point machines
IEEE Transactions on Circuits and Systems for Video Technology
Multimedia Search Without Visual Analysis: The Value of Linguistic and Contextual Information
IEEE Transactions on Circuits and Systems for Video Technology
Extracting semantics from audio-visual content: the final frontier in multimedia retrieval
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
Multimedia content is usually complex and may contain many semantically meaningful elements interrelated to each other. Therefore to understand the high-level semantic meanings of the content, such interrelations need to be learned and exploited to further improve the search process. We introduce our ideas on how to enable automatic construction of semantic context by learning from the content. Depending on the targeted source of content, representation schemes for its semantic context can be constructed by learning from data. In the target representation scheme, metadata is divided into three levels: low, mid, and high levels. By using the proposed scheme, high-level features are derived out of the mid-level features. In order to explore the hidden interrelationships between mid-level and the high-level terms, a Bayesian network model is built using from a small amount of training data. Semantic inference and reasoning is then performed based on the model to decide the relevance of a video.