Exploring semantic dependencies for scalable concept detection

Semantic concept detection from multimedia features enables high-level access to multimedia content. While constructing robust detectors is feasible for concepts with sufficient training samples, concepts with fewer training samples are hard to train efficiently. Comparable performance may be possible if the dependence of these concepts on the ones that can be robustly modeled is exploited. In this paper we show this phenomenon using the TREC Video 2002 Corpus as a test bed. Using a basic set of 12 semantic concepts modeled with support vector machines, we predict presence of 4 other concepts. We then compare the performance of these predictors with direct SVM models for these 4 concepts and observe improvements of up to 150% in average precision.

[1]  Ioannis Pitas,et al.  On the stability of support vector machines for face detection , 2002, Proceedings. International Conference on Image Processing.

[2]  John R. Smith,et al.  Active selection for multi-example querying by content , 2003, 2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698).

[3]  John R. Smith,et al.  A framework for moderate vocabulary semantic visual concept detection , 2003, 2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698).

[4]  Thomas S. Huang,et al.  Factor graph framework for semantic video indexing , 2002, IEEE Trans. Circuits Syst. Video Technol..

[5]  James Ze Wang,et al.  Learning-based linguistic indexing of pictures with 2--d MHMMs , 2002, MULTIMEDIA '02.

[6]  Thomas S. Huang,et al.  Relevance feedback: a power tool for interactive content-based image retrieval , 1998, IEEE Trans. Circuits Syst. Video Technol..

[7]  Jeffrey Scott Vitter,et al.  CAMEL: concept annotated image libraries , 2001, IS&T/SPIE Electronic Imaging.

[8]  Sharad Mehrotra,et al.  Similarity Search Using Multiple Examples in MARS , 1999, VISUAL.

[9]  John R. Smith,et al.  Multimedia semantic indexing using model vectors , 2003, 2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698).