Abstract
The multimodal data usually contain complementary, correlated and redundant information. Thus, multimodal fusion is useful for many multisensor applications. Here, a novel multimodal fusion algorithm is proposed, which is referred to as “MultiFusion.” The approach adopts a boosting structure where the atomic event is considered as the fusion unit. The correlation of multimodal data is used to form an overall classifier in each iteration. Moreover, by adopting the Adaboost-like structure, the overall fusion performance is improved. Both the simulation experiment and the real application show the effectiveness of the MultiFusion approach. Our approach can be applied in different multimodal applications to exploit the multimedia data characteristics and improve the performance.
- Aly, R. and Hiemstra, D. 2009. Concept detectors: how good is good enough? In Procedings of the ACM International Conference on Multimedia. 233--242. Google ScholarDigital Library
- Amores, J., Sebe, N., Radeva, P., Gevers, T., and Smeulders, A. W. M. 2004. Boosting contextual information in content-based image retrieval. In Proceedings of the 6th ACM SIGMM International Workshop on Multimedia Information Retrieval. 31--38. Google ScholarDigital Library
- Atrey, P. K., Kankanhalli, M. S., and Jain, R. 2006. Information assimilation framework for event detection in multimedia surveillance systems. Multimedia Syst. 12, 3, 239--253.Google ScholarDigital Library
- Barbu, C., Iqbal, R. T., and Peng, J. 2005. Classifier fusion using shared sampling distribution for boosting. In Proceedings of the International Conference on Data Mining. 34--41. Google ScholarDigital Library
- Contreras, F. M., Urunuela, C. O., and del Rincon, J. M. 2009. Adaboost multiple feature selection and combination for face recognition. In Proceedings of the Iberian Conference on Pattern Recognition and Image Analysis. 338--345. Google ScholarDigital Library
- Dasarathy, B. V. 1994. Decision Fusion. Computer Society Press. Google ScholarDigital Library
- Freund and Schapire. 1997. A decision-theoretic generalization of on-line learning and an application to boosting. J. Compu. Syst. Sci. 55. Google ScholarDigital Library
- Freund, Y. and Schapire, R. E. 1996. Experiments with a new boosting algorithm. In Proceedings of the International Conference on Machine Learning. 148--156.Google Scholar
- Guo, G. and Zhang, H. 2001. Boosting for fast face recognition. Tech. rep. MSR-TR-2001-16, Microsoft Research.Google Scholar
- Guo, W., Qi, Q., Jiang, L., and Zhang, P. 2008. A new method of sar image target recognition based on adaboost algorithm. In Proceedings of the IEEE International Geoscience and Remote Sensing Symposium.Google Scholar
- Iyengar, G., Nock, H., and Neti, C. 2003. Audio-visual synchrony for detection of monologues in video archives. In Proceedings of the IEEE International Conference on Multimedia and Expo 1, 329--332. Google ScholarDigital Library
- Joshi, D., Naphade, M. R., and Natsev, A. 2007. Semantics reinforcement and fusion learning for multimedia streams. In Proceedings of the International Conference on Image and Video Retrieval. 309--316. Google ScholarDigital Library
- Maghooli, K. and Moin, M. S. 2004. A new approach on multimodal biometrics based on combining neural networks using adaboost. In Biometric Authentication. Lecture Notes in Computer Science Vol. 3087 332--341.Google Scholar
- Parikh, D., Kim, M. T., Oagaro, J., Mandayam, S., and Polikar, R. 2004. Combining classifiers for multisensor data fusion. In Proceedings of the International Conference on Systems, Man and Cybernetics. 1232--1237.Google Scholar
- Pickering, M. J., Ruger, S. M., and Sinclair, D. 2002. Video retrieval by feature learning in key frames. In Proceedings of the International Conference on Image and Video Retrieval. 309--317. Google ScholarDigital Library
- Polikar, R. 2006. Ensemble based systems in decision making. IEEE Circu. Syst. Maga. 6, 3, 21--45.Google ScholarCross Ref
- Polikar, R., Upda, L., Upda, S. S., and Honavar, V. 2001. Learn++: an incremental learning algorithm for supervised neural networks. IEEE Trans. Syst. Man, Cyber. Part C 31, 4, 497--508. Google ScholarDigital Library
- Schapire, R. E. 1990. The strength of weak learnability. Mach. Learn. 5, 2. Google ScholarDigital Library
- Tieu, K. and Viola, P. 2004. Boosting image retrieval. Int. J. Comput. Vis. 56, 1--2, 17--36. Google ScholarDigital Library
- Xue, F. and Ding, X. 2006. 3D+2D face localization using boosting in multi-modal feature space. In Proceedings of the International Conference on Pattern Recognition. III: 499--502. Google ScholarDigital Library
- Zhang, L., Li, M. J., and Zhang, H. J. 2002. Boosting image orientation detection with indoor vs. outdoor classification. In Proceedings of the IEEE Workshop on Applications of Computer Vision. 95--99. Google ScholarDigital Library
- Zhang, T. and Yu, B. 2005. Boosting with early stopping: Convergence and consistency. Ann. Stat. 33, 1538--1579.Google ScholarCross Ref
Index Terms
- MultiFusion: A boosting approach for multimedia fusion
Recommendations
On Comparing Early and Late Fusion Methods
Advances in Computational IntelligenceAbstractThis paper presents a theoretical comparison of early and late fusion methods. An initial discussion on the conditions to apply early or late (soft or hard) fusion is introduced. The analysis show that, if large training sets are available, early ...
A multilevel information fusion approach for visual quality inspection
For visual quality inspection systems to be applicable in industrial settings, it is mandatory that they are highly flexible, robust and accurate. In order to improve these characteristics a multilevel information fusion approach is presented. A first ...
Estimation and decision fusion: A survey
Data fusion has been applied to a large number of fields and the corresponding applications utilize numerous mathematical tools. This survey focuses on some aspects of estimation and decision fusion. In estimation fusion, we discuss the development of ...
Comments