A nonparametric Bayesian learning model using accelerated variational inference and feature selection
- 79 Downloads
Developing effective machine learning methods for multimedia data modeling continues to challenge computer vision scientists. The capability of providing effective learning models can have significant impact on various applications. In this work, we propose a nonparametric Bayesian approach to address simultaneously two fundamental problems, namely clustering and feature selection. The approach is based on infinite generalized Dirichlet (GD) mixture models constructed through the framework of Dirichlet process and learned using an accelerated variational algorithm that we have developed. Furthermore, we extend the proposed approach using another nonparametric Bayesian prior, namely Pitman–Yor process, to construct the infinite generalized Dirichlet mixture model. Our experiments, which were conducted through synthetic data sets, the clustering analysis of real-world data sets and a challenging application, namely automatic human action recognition, indicate that the proposed framework provides good modeling and generalization capabilities.
KeywordsInfinite mixtures Variational Bayes Generalized Dirichlet Feature selection Human action recognition
Funding was provided by National Natural Science Foundation of China (Grant No. 61502183), The Scientific Research Funds of Huaqiao University (Grant No. 600005-Z15Y0016) and Promotion Program for Young and Middle-aged Teacher in Science and Technology Research of Huaqiao University (Grant No. ZQN-PY510).
- 4.Bouguila N (2007) Spatial color image databases summarization. In: Proc. of the IEEE international conference on acoustics, speech and signal processing (ICASSP 2007), vol 1, pp I-953–I-956Google Scholar
- 5.Bouguila N, Ziou D (2004a) Improving content based image retrieval systems using finite multinomial Dirichlet mixture. In: Proc. of the 14th IEEE signal processing society workshop on machine learning for signal processing, pp 23–32Google Scholar
- 6.Bouguila N, Ziou D (2004b) A powerful finite mixture model based on the generalized Dirichlet distribution: unsupervised learning and applications. In: Proc. of the 17th international conference on pattern recognition (ICPR 2004), vol 1, pp 280–283 Vol 1Google Scholar
- 14.Kuehne H, Jhuang H, Garrote E, Poggio T, Serre T (2011) HMDB: a large video database for human motion recognition. In: Proc. of the international conference on computer vision (ICCV), pp 2556–2563Google Scholar
- 15.Kurihara K, Welling M, Vlassis N (2006) Accelerated variational Dirichlet process mixtures. In: Proc. of advances in neural information processing systems (NIPS)Google Scholar
- 17.Laptev I, Marszalek M, Schmid C, Rozenfeld B (2008) Learning realistic human actions from movies. In: Proc. of IEEE conference on computer vision and pattern recognition (CVPR), pp 1–8Google Scholar
- 21.Nguyen NT, Zheng G, Han Z, Zheng R (2011) Device fingerprinting to enhance wireless security using nonparametric Bayesian method. In: Proc. of the IEEE conference on INFOCOM, pp 1404–1412Google Scholar
- 24.Shyr A, Darrell T, Jordan M, Urtasun R (2011) Supervised hierarchical Pitman–Yor process for natural scene segmentation. In: Proc. of the 2011 IEEE conference on computer vision and pattern recognition (CVPR), pp 2281–2288Google Scholar
- 26.Sudderth EB, Jordan MI (2008) Shared segmentation of natural scenes using dependent Pitman-Yor processes. In: Proc. of Advances in Neural Information Processing Systems (NIPS), pp 1585–1592Google Scholar
- 27.Teh YW (2006) A hierarchical Bayesian language model based on Pitman-Yor processes. In: Proc. of the 21st International Conference on Computational Linguistics and the 44th Annual Meeting of the Association for Computational Linguistics, ACL-44, pp 985–992Google Scholar
- 30.Wang T, Hammoud R, Zhu Z (2014) Ground-based activity recognition at distance and behind wall. In: Proc. of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp 231–236Google Scholar