Abstract : We firstly propose continuous probabilistic latent semantic analysis (PLSA) to model continuous quantity. In addition, corresponding Expectation-Maximization (EM) algorithm is derived to determine the model parameters. Furthermore, we present a hybrid framework which employs continuous PLSA to model visual features of images in generative learning stage and uses ensembles of classifier chains to classify the multi-label data in discriminative learning stage. Since the framework combines the advantages of generative and discriminative learning, it can predict semantic annotation precisely for unseen images. Finally, we conduct a series of experiments on a standard Corel dataset. The experiment results show that our approach outperforms many state-of-the-art approaches.
Zhongzhi Shi; David Leake; Sunil Vadera. 7th International Conference on Intelligent Information Processing (IIP), Oct 2012, Guilin, China. Springer, IFIP Advances in Information and Communication Technology, AICT-385, pp.347-356, 2012, Intelligent Information Processing VI. 〈10.1007/978-3-642-32891-6_43〉
https://hal.inria.fr/hal-01524985
Contributeur : Hal Ifip
<>
Soumis le : vendredi 19 mai 2017 - 10:43:39
Dernière modification le : mercredi 17 octobre 2018 - 16:30:04
Document(s) archivé(s) le : mardi 22 août 2017 - 00:58:48
Zhixin Li, Weizhong Zhao, Zhiqing Li, Zhiping Shi. Automatic Image Annotation and Retrieval Using Hybrid Approach. Zhongzhi Shi; David Leake; Sunil Vadera. 7th International Conference on Intelligent Information Processing (IIP), Oct 2012, Guilin, China. Springer, IFIP Advances in Information and Communication Technology, AICT-385, pp.347-356, 2012, Intelligent Information Processing VI. 〈10.1007/978-3-642-32891-6_43〉. 〈hal-01524985〉
