基于概率主题模型的图像分类和标注的研究
|
摘要
图像分类和标注是计算机视觉、机器学习领域的重要研究内容。它们是自动获取图像语义信息的重要手段,具有广泛的应用,然而也面临着一定的挑战。概率图模型和变分推理是一个新型的机器学习框架,是处理不确定性和复杂性问题的有力工具,被广泛应用于计算机视觉和自然语言处理等领域。主题模型源于文本处理,主题模型的优势在于：一、可对数据降维,二、可使用学习到的主题特征(中层特征)代替原始的底层特征来分类,从而缩小高层类别概念与底层图像特征的距离。本文注意到现有很多图像分类和标注方法应用到真实图像中存在一定的局限性,因而提出了一系列新颖的、用于图像分类和标注的概率主题模型,期望有效提高图像分类和标注的性能。具体创新工作如下：
1.提出了一个集成的、有监督的概率主题模型。很多真实的图像数据在形式上或底层特征上表现为类内相似度很小,而类问相似度却很大。一般来说,对于这样复杂的图像,一个规则很难拟合数据和类标之间的关系。我们认为这样的数据应该存在多个分类规则,这就涉及到了集成学习问题,集成学习的优势是组合多个弱分类器构成一个强分类器。目前,基于主题模型的图像分类工作已有很多,它们大都为所有数据构建一个分类规则。因此,本文在主题模型中引入了集成学习的思想,构建了一个可以兼顾两个方法优点的分类模型,并在两个真实数据集上证明了提出模型的合理性。
2.提出了一个多视图的、有监督的概率主题模型。在计算机视觉中多视图特征是很容易获得的。对于复杂图像,一般来说一个特征是很难具有足够辨识度的。显然,多种特征可以提供较高的辨识度,辨识度越高,分类也就越容易。现存的主题模型大都关注单视图特征,若想使用多特征信息,只能使用多个特征的联合特征。由于特征所在不同的尺度空间,简单拼接常常是不合理的。为此,本文构建了一个多视图的有监督的概率主题模型,并在两个真实数据集上证明了提出模型的有效性。
3.提出了一个用类别信息促进图像标注的概率主题模型。考虑到类标信息对图像标注来说是很有价值的,一旦我们确定了图像类别,标准词汇的范围将会缩小,标注的误差就会减少。并且,在计算机视觉的很多任务中,相比图像的标注信息,类标常常是很容易获得的。因此本文构建了一个用类别信息标注图像的概率主题模型。最后在两个真实数据集上证明了该模型的有效性。
4.提出了同时做图像分类和标注的概率主题模型。考虑到不同的类别常常联系着不同的标注词汇,一旦确定了图像的类别,标注词汇的范围将会缩小,标注的误差就会减少。反之亦然,一旦知道了标注信息,类标词的范围也会缩小,即类标和标注之问可相互提供有价值的信息。可见,图像分类和标注是有关联的,不仅可以同时做,而且可以互相促进。受此启发,本文构建了一个新颖的同时做图像分类和标注的概率主题模型。两个真实数据集上的实验结果表明了我们方法的分类性能与相比较方法中的最好性能相差不多,而标注性能有了很大提高,也表明了提出模型的合理性。
In computer vision, and machine learning, image classification and annotation have been important, and are important methods obtaining image sementic. They have been applied widely, however, they also meet big challenge. Probabilistic graphical model and variational inference is a new machine learning framework, and is a good method applied to uncertain problem or complex problem. It is applied in computer vision and natural language processing widely. Topic model is original from text processing. An advantage of topic model is data dimension reduction, and another one is it can use the latent topic feature for classification, so that it can narrow the distance between category concept and features. The paper considers that the existing image classification methods and annotation methods are limited in the complex real images, and so we propose a series of probalistic topic models for image classification and annotation. We aim to improve the performance of image classification and annotation. In particular, the innovation is shown as follows:
1. The paper proposes an emsemble supervised probabilistic topic model. In the real word, the similarity of intro-class is big, the similarity of inter-class is small. In general, only one classification "criterion"(classifier) is difficult to fit the relation between the complex image and class label. For the complex image data, we think it should exist multiple criteria. It is the problem which ensemble methods can resolve. Ensemble methods can combine multiple weak classifiers to construct a strong classifier. Much work based topic model for image classification has been done now. They mostly construct a single classification criterion for all training data. Our motivating intuition is, here, introducing the ensemble classification idea to topic model, so as to construct a model which can combine the merits of the two kinds of methods. The experimental results on two real image datasets show the effectiveness of the proposed model.
2. The paper proposes a multi-view supervised probabilistic topic model. In computer vision, multiple view features are obtained easily. For complex images, a single feature can not have enough recognition. Obviously, multiple features have better recognition capability than a single feature. The better recognition, the easier classification. The existing topic models for classication mostly focus on single view of features. If they try to consider the information of multiple features, they can only use the union feature of multiple features. Generally, however, simple union of multi-features is not reasonable because of different scales among the views. Thus, motivated by this, we try to construct a probabilistic topic model for classifying data with multi-view features. The experimental results show the effectiveness of the proposed model.
3. The paper proposes a probabilistic topic model which can impove image annotation using category information. We consider that category information can provide certain evidence, or valuable information for image annotation. Once the category of the images was ascertained, it is equivalent to reduce some uncertainty of annotation. More over, category information is obtained easily comparing with annotation information in computer vision. It motivates us to construct a probabilistic topic model which can impove image annotation using category information. The experimental results show the effectiveness of the proposed model.
4. The paper proposes a novel probabilistic topic model for simultaneous image classification and annotation. We draw on the fact that once the category of an image is ascertained, the scope of annotation words for the image can be narrowed. Meanwhile, the probability of generating irrelevant annotation words can be reduced, and vice versa. As such, we think not only these two tasks of image classification and image annotation can be performed simutaneously, but also they can be implemented in ways that improve one another. Based on this intuition, we propose a novel probabilistic topic model for simultaneous image classification and annotation. The performance of the proposed model is demonstrated on two real-word datasets, and the results show that our model provides a competitive classification performance with several benchmark classification models, while it shows a better annotation performance than other benchmark annotation models. The experimental results also show the rationality of the proposed model.
1.提出了一个集成的、有监督的概率主题模型。很多真实的图像数据在形式上或底层特征上表现为类内相似度很小,而类问相似度却很大。一般来说,对于这样复杂的图像,一个规则很难拟合数据和类标之间的关系。我们认为这样的数据应该存在多个分类规则,这就涉及到了集成学习问题,集成学习的优势是组合多个弱分类器构成一个强分类器。目前,基于主题模型的图像分类工作已有很多,它们大都为所有数据构建一个分类规则。因此,本文在主题模型中引入了集成学习的思想,构建了一个可以兼顾两个方法优点的分类模型,并在两个真实数据集上证明了提出模型的合理性。
2.提出了一个多视图的、有监督的概率主题模型。在计算机视觉中多视图特征是很容易获得的。对于复杂图像,一般来说一个特征是很难具有足够辨识度的。显然,多种特征可以提供较高的辨识度,辨识度越高,分类也就越容易。现存的主题模型大都关注单视图特征,若想使用多特征信息,只能使用多个特征的联合特征。由于特征所在不同的尺度空间,简单拼接常常是不合理的。为此,本文构建了一个多视图的有监督的概率主题模型,并在两个真实数据集上证明了提出模型的有效性。
3.提出了一个用类别信息促进图像标注的概率主题模型。考虑到类标信息对图像标注来说是很有价值的,一旦我们确定了图像类别,标准词汇的范围将会缩小,标注的误差就会减少。并且,在计算机视觉的很多任务中,相比图像的标注信息,类标常常是很容易获得的。因此本文构建了一个用类别信息标注图像的概率主题模型。最后在两个真实数据集上证明了该模型的有效性。
4.提出了同时做图像分类和标注的概率主题模型。考虑到不同的类别常常联系着不同的标注词汇,一旦确定了图像的类别,标注词汇的范围将会缩小,标注的误差就会减少。反之亦然,一旦知道了标注信息,类标词的范围也会缩小,即类标和标注之问可相互提供有价值的信息。可见,图像分类和标注是有关联的,不仅可以同时做,而且可以互相促进。受此启发,本文构建了一个新颖的同时做图像分类和标注的概率主题模型。两个真实数据集上的实验结果表明了我们方法的分类性能与相比较方法中的最好性能相差不多,而标注性能有了很大提高,也表明了提出模型的合理性。
In computer vision, and machine learning, image classification and annotation have been important, and are important methods obtaining image sementic. They have been applied widely, however, they also meet big challenge. Probabilistic graphical model and variational inference is a new machine learning framework, and is a good method applied to uncertain problem or complex problem. It is applied in computer vision and natural language processing widely. Topic model is original from text processing. An advantage of topic model is data dimension reduction, and another one is it can use the latent topic feature for classification, so that it can narrow the distance between category concept and features. The paper considers that the existing image classification methods and annotation methods are limited in the complex real images, and so we propose a series of probalistic topic models for image classification and annotation. We aim to improve the performance of image classification and annotation. In particular, the innovation is shown as follows:
1. The paper proposes an emsemble supervised probabilistic topic model. In the real word, the similarity of intro-class is big, the similarity of inter-class is small. In general, only one classification "criterion"(classifier) is difficult to fit the relation between the complex image and class label. For the complex image data, we think it should exist multiple criteria. It is the problem which ensemble methods can resolve. Ensemble methods can combine multiple weak classifiers to construct a strong classifier. Much work based topic model for image classification has been done now. They mostly construct a single classification criterion for all training data. Our motivating intuition is, here, introducing the ensemble classification idea to topic model, so as to construct a model which can combine the merits of the two kinds of methods. The experimental results on two real image datasets show the effectiveness of the proposed model.
2. The paper proposes a multi-view supervised probabilistic topic model. In computer vision, multiple view features are obtained easily. For complex images, a single feature can not have enough recognition. Obviously, multiple features have better recognition capability than a single feature. The better recognition, the easier classification. The existing topic models for classication mostly focus on single view of features. If they try to consider the information of multiple features, they can only use the union feature of multiple features. Generally, however, simple union of multi-features is not reasonable because of different scales among the views. Thus, motivated by this, we try to construct a probabilistic topic model for classifying data with multi-view features. The experimental results show the effectiveness of the proposed model.
3. The paper proposes a probabilistic topic model which can impove image annotation using category information. We consider that category information can provide certain evidence, or valuable information for image annotation. Once the category of the images was ascertained, it is equivalent to reduce some uncertainty of annotation. More over, category information is obtained easily comparing with annotation information in computer vision. It motivates us to construct a probabilistic topic model which can impove image annotation using category information. The experimental results show the effectiveness of the proposed model.
4. The paper proposes a novel probabilistic topic model for simultaneous image classification and annotation. We draw on the fact that once the category of an image is ascertained, the scope of annotation words for the image can be narrowed. Meanwhile, the probability of generating irrelevant annotation words can be reduced, and vice versa. As such, we think not only these two tasks of image classification and image annotation can be performed simutaneously, but also they can be implemented in ways that improve one another. Based on this intuition, we propose a novel probabilistic topic model for simultaneous image classification and annotation. The performance of the proposed model is demonstrated on two real-word datasets, and the results show that our model provides a competitive classification performance with several benchmark classification models, while it shows a better annotation performance than other benchmark annotation models. The experimental results also show the rationality of the proposed model.
引文
[1]Li Li Jia and Li Fei Fei, "What, where and who? Classifying events by scene and object recognition," in 2007 IEEE 11th International Conference on Computer Vision, October 14,2007-October 21,2007, Rio de Janeiro, Brazil,2007.
[2]钟义信,机器知行学原理.北京：北京出版社,2007.
[3]Bishop Christopher M., Pattern Recognition and Machine Learning second ed. Secaucus, NJ, USA:Springer-Verlag New York,2006.
[4]Tarek Habib, Jordi Inglada, Gregoire Mercier, et al., "Speeding up support vector machine (SVM) image classification by a kernel series expansion," in 2008 IEEE International Conference on Image Processing, October 12,2008-October 15,2008, San Diego, CA, United states,2008, pp.865-868.
[5]Begum Demir and Erturk Sarp, "Improving SVM classification accuracy using a hierarchical approach for hyperspectral images," in 2009 IEEE International Conference on Image Processing, November 7,2009-November 12,2009, Cairo, Egypt, 2009, pp.2849-2852.
[6]沈新宇,许宏丽,官腾飞,”基于直推式支持向量机的图像分类算法,”计算机应用,vol.6,2007.
[7]Bosch Anna, Zierssman Andrew and Munoz, Xavierl, "Image classification using random forests and ferns," in 2007 IEEE 11th International Conference on Computer Vision, October 14,2007-October 21,2007, Rio de Janeiro, Brazil,2007.
[8]Li, Yan and Li, Lei, "A novel split and merge EM algorithm for Gaussian mixture model," in 5th International Conference on Natural Computation, August 14, 2009-August 16,2009, Tianjian, China,2009, pp.479-483.
[9]Vailaya Aditya, Figueiredo Mario A. T., JainAnil K., et al. "Image classification for content-based indexing," IEEE Transactions on Image Processing, vol. 10,2001, pp.117-130.
[10]Hofmann Thomas, "Unsupervised learning by probabilistic Latent Semantic Analysis," Machine Learning, vol.42,2001, pp.177-196.
[11]Blei David M., Ng Andrew Y. and Jordan Michael I., "Latent Dirichlet allocation," Journal of Machine Learning Research, vol.3,2003, pp.993-1022.
[12]Sivic Josef, Russell Bryan C.,Efros Alexei A., et al, "Discovering objects and their location in images," in Proceedings-10th IEEE International Conference on Computer Vision, October 17,2005-October 20,2005, Beijing, China, 2005, pp.370-377.
[13]Quelhas P., Monay F., Odobez, J.-M. Gatica-Perez D., et al., "Modeling scenes with local descriptors and latent aspects," in Proceedings-10th IEEE International Conference on Computer Vision, October 17,2005-October 20,2005, Beijing, China,2005, pp.883-890.
[14]Bosch Anna, Zisserman Andrew, Munoz Xavier, "Scene classification via pLSA," in 9th European Conference on Computer Vision, May 7,2006-May 13,2006, Graz, Austria,2006, pp.517-530.
[15]Li Fei-Fei and Perona Pietro, "A bayesian hierarchical model for learning natural scene categories," in 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 20,2005-June 25,2005, San Diego, CA, United states,2005, pp.524-531.
[16]Wang Chong, Blei David and Li Fei-Fei, "Simultaneous image classification and annotation," in 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, June 20,2009-June 25,2009, Miami, FL, United states,2009, pp.1903-1910.
[17]Yang Yi, Wu Fei, Nie Feiping, et al., "Web and personal image annotation by mining label correlation with relaxed visual graph embedding," IEEE Transactions on Image Processing, vol.21,2012, pp.1339-1351.
[18]Li Xirong, Gavves Efstratios, Snoek Cees G.M., et al., "Personalizing automated image annotation using cross-entropy," in 19th ACM International Conference on Multimedia ACM Multimedia 2011, November 28,2011-December 1, 2011, Scottsdale, AZ, United states,2011, pp.233-242.
[19]Wang Changhu, Jing Feng, Zhang Lei, et al., "Content-based image annotation refinement," in 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 17,2007-June 22,2007, Minneapolis, MN, United states,2007.
[20]Karadag Ozge Oztimur and Vural Fatos T. Yarman, "HANOLISTIC:A hierarchical automatic image annotation system using holistic approach," in 2009 IEEE Conference on Computer Vision and Pattern Recognition, June 20,2009-June 25, 2009, Miami, FL, United states,2009, pp.16-21.
[21]Zhao Yufeng, Zhao Yao, Zhu Zhenfeng, "Pseudo relevance feedback for searech-based image annotation," Chinese Journal of Electronics, vol.17,2008, pp. 471-474.
[22]Wang Hua, Huang Heng, Ding Chris, "Image annotation using multi-label correlated Green's function," in 12th International Conference on Computer Vision, September 29,2009-October 2,2009, Kyoto, Japan,2009, pp.2029-2034.
[23]Mueen A.l,Zainuddin R., Baba M. Sapiyan, "Automatic multilevel medical image annotation and retrieval," Journal of Digital Imaging, vol.21,2008, pp. 290-295.
[24]Barnard K. and Forsyth D., "Learning the semantics of words and pictures," in 8th International Conference on Computer Vision, July 9,2001-July 12, 2001, Vancouver, BC, United states,2001, pp.408-415.
[25]Gao Yan-Yu, Yin Yi-Xin, Uozumi Takashi, "A hierarchical image annotation method based on SVM and semi-supervised EM," Zidonghua Xuebao/Acta Automatica Sinica, vol.36,2010, pp.960-967.
[26]Qi Xiaojun and Han Yutao, "Incorporating multiple SVMs for automatic image annotation," Pattern Recognition, vol.40,2007, pp.728-741.
[27]Han Yutao, Qi Xiaojun, "A complementary SVMs-based image annotation system," in IEEE International Conference on Image Processing 2005, ICIP 2005, September 11,2005-September 14,2005, Genova, Italy,2005, pp.1185-1188.
[28]Zhao Yufeng, Zhao Yao, Zhu Zhenfeng, "A novel image annotation scheme based on neural network," in 8th International Conference on Intelligent Systems Design and Applications, November 26,2008-November 28,2008, Kaohsiung, Taiwan, 2008, pp.644-647.
[29]王科平,”自动图像标注的关键技术研究,”博士论文,北京邮电大学,2011.
[30]Feng Songhe, Xu De, Li Bing, "Automatic region-based image annotation using an improved multiple-instance learning algorithm," Chinese Journal of Electronics, vol.17,2008,pp.43-47.
[31]Yang Changbo, Dong Ming and Hua Jing, "Region-based image annotation using asymmetrical support vector machine-based multiple-instance learning," in 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 17,2006-June 22,2006, New York, NY, United states,2006, pp.2057-2063.
[32]Putthividhya D., Attias H.T. and Nagarajan, S.S., "Supervised topic model for automatic image annotation," in 2010 IEEE International Conference on Acoustics Speech, and Signal Processing, March 14,2010-March 19,2010, Dallas, TX, United states,2010, pp.1894-1897.
[33]Blei David M. and McAuliffe Jon D., "Supervised topic models," in 21st Annual Conference on Neural Information Processing Systems, December 3,2007 December 6,2007, Vancouver, BC, Canada,2009.
[34]Blei David M. and Jordan Michael I., "Modeling Annotated Data," in Proceedings of the Twenty-Sixth Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, July 28,2003-August 1,2003, Toronto, Ont., Canada,2003, pp.127-134.
[35]Jeon J., Lavrenko V. and Manmatha R., "Automatic Image Annotation and Retrieval using Cross-Media Relevance Models," in Proceedings of the Twenty-Sixth Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, July 28,2003-August 1,2003, Toronto, Ont., Canada,2003, pp. 119-126.
[36]Lavrenko V, Manmatha and R. J. Jeon, "A model for leaning the semantics of pictures," presented at the Neural Information Processing Systems (NIPS 2004).
[37]Putthividhya Duangmanee, Attias Hagai T. and Nagarajan Srikantan S., "Topic regression multi-modal Latent Dirichlet Allocation for image annotation," in 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 13,2010-June 18,2010, San Francisco, CA, United states,2010, pp. 3408-3415.
[38]bag of words. Available: http://en.wikipedia.org/wiki/Bag_of_words_model
[39]Zhang Yin, Jin Rong and Zhou Zhi-Hua, "Understanding bag-of-words model: A statistical framework," International Journal of Machine Learning and Cybernetics, vol.1,2010, pp.43-52.
[40]Wu Lei, Hoi, Steven C. H. and Yu Nenghai, "Semantics-preserving bag-of-words models and applications," IEEE Transactions on Image Processing, vol. 19,2010, pp.1908-1920.
[41]Li Teng, Mei Tao, Kweon In-So, Hua Xian-Sheng, "Contextual bag-of-words for visual categorization," IEEE Transactions on Circuits and Systems for Video Technology, vol.21, pp.381-392,2011.
[42]Lowe David G.., "Object recognition from local scale-invariant features," in Proceedings of the 1999 7th IEEE International Conference on Computer Vision, September 20,1999-September 27,1999, Kerkyra, Greece,1999, pp.1150-1157.
[43]Wu Jianxin, Rehg Jim M., "CENTRIST:A visual descriptor for scene categorization," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 33,2011,pp.1489-1501.
[44]Wu Jianxin and Rehg James M., "Beyond the Euclidean distance:Creating effective visual codebooks using the histogram intersection kernel," in 12th International Conference on Computer Vision, September 29,2009-October 2,2009, Kyoto, Japan,2009, pp.630-637.
[45]Bayesian network. Available: http://en.wikipedia.org/wiki/Bayesian_network
[46]Liang Lin, Wen Fang, Xu Ying-Qing, et al., "Accurate face alignment using shape constrained Markov network," in 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 17,2006-June 22,2006, New York, NY, United states,2006, pp.1313-1320.
[47]Barrat Sabine and Tabbone Salvatore, "Modeling, classifying and annotating weakly annotated images using Bayesian network," Journal of Visual Communication and Image Representation, vol.21,2010, pp.355-363.
[48]Alam Fahim Irfanl and Gondra Iker, "A Bayesian network-based tunable image segmentation algorithm for object recognition," in 11th IEEE International Symposium on Signal Processing and Information Technology, December 14,2011 December 17,2011, Bilbao, Spain,2011, pp.011-016.
[49]Dewar Michael, Wiggins Chris and Wood Frank, "Inference in hidden Markov models with explicit state duration distributions," IEEE Signal Processing Letters, vol.19,2012, pp.235-238.
[50]Markov Random Field. Available: http://en.wikipedia.org/wiki/Markoy_random_field
[51]Galecki Andrzej T., Ten Have Thomas R. and Molenberghs Geert, "Simple and fast alternative to the EM algorithm for incomplete categorical data and latent class models," Computational Statistics and Data Analysis, vol.35,2001, pp.265-281.
[52]Teh Yee Whye, Kurihara Kenichi and Welling Max, "Collapsed variational inference for HDP," in 21st Annual Conference on Neural Information Processing Systems, December 3,2007-December 6,2007, Vancouver, BC, Canada,2009.
[53]Wang Chong and Blei David M., "Variational inference for the nested Chinese restaurant process," in 23rd Annual Conference on Neural Information Processing Systems, December 7,2009-December 10,2009, Vancouver, BC, Canada, 2009, pp.1990-1998.
[54]Liao Chih-Pin and Chien Jen-Tzung, "Variational inference for conditional random fields," in 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, March 14,2010-March 19,2010, Dallas, TX, United states,2010, pp.2002-2005.
[55]Monte Carlo. Available:http://en.wikipedia.org/wiki/Monte_Carlo_method
[56]Jordan Michael I., "An introduction to variational methods for graphical models," Machine Learning, vol.37, pp.183-233,1999.
[57]maximum entropy. Available: http://en.wikipedia.org/wiki/Maximum_entropy
[58]Kullback-Leibler divergence. Available: http://en.wikipedia.org/wiki/Kullback%E2%80%93Leibler_divergence
[59]Tanaka Toshiyuki, "Information geometry of mean field theory," IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, 1996, vol. E79-A, pp.709-715.
[60]mean field theory. Available: http://en.wikipedia.org/wiki/Mean_field_theory
[61]Generative Model. Available: http://en.wikipedia.org/wiki/Generative_model
[62]Dirichlet Distribution. Available: http://en.wikipedia.org/wiki/Pirichlet_distribution
[63]Multinomial Distribution. Available: http://en.wikipedia.org/wiki/Multinomial_distribution
[64]Simplex. Available:http://en.wikipedia.org/wiki/Simplex
[65]Gamma function. Available:http://en.wikipedia.org/wiki/Gamma_function
[66]Nocedal Jorge and Wright Stephen J., Numerical Optimization. Berlin, Germany:Springer,2006.
[67]Russell Bryan C., Torralba Antonio, Murphy Kevin P., et al., "LabelMe:A database and web-based tool for image annotation," International Journal of Computer Vision, vol.77,2008, pp.157-173.
[68]Cao Liangliang, Li Fei-Fei, "Spatially coherent latent topic model for concurrent segmentation and classification of objects and scenes," in 2007 IEEE 11th International Conference on Computer Vision, October 14,2007-October 21,2007, Rio de Janeiro, Brazil,2007.
[69]Nishii Ryuei and Eguchi Shinto, "Supervised image classification based on AdaBoost with contextual weak classifiers," in 2004 IEEE International Geoscience and Remote Sensing Symposium Proceedings:Science for Society:Exploring and Managing a Changing Planet. September 20,2004-September 24,2004, Anchorage, AK, United states,2004, pp.1467-1470.
[70]Gao Feng and Ai Haizhou, "Face age classification on consumer images with gabor feature and fuzzy LDA method," in 3rd International Conference on Advances in Biometrics, June 2,2009-June 5,2009, Alghero, Italy,2009, pp.132-141.
[71]Chen Kai and Jin Yang, "An ensemble learning algorithm based on Lasso selection," in 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, October 29,2010-October 31,2010, Xiamen, China,2010, pp. 617-620.
[72]Liuxia and Lisu, "Research on Face Detection Classifier using an improved Adboost algorithm," in International Symposium on Computer Science and Computational Technology, December 20,2008-December 22,2003, Shanghai, China, 2008, pp.78-81.
[73]Wang Gang, Hoiem Derek and Forsyth David, "Building text features for object image classification," in 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, June 20,2009-June 25,2009, Miami, FL, United states,2009, pp.1367-1374.
[74]Li Xiaoxu, Zeng Junyu,Wang Xiaojie, et al., "Mixture of softmax sLDA," in 11th IEEE International Conference on Data Mining, ICDM 2011, December 11, 2011-December 14,2011, Vancouver, BC, Canada,2011, pp.1164-1169.
[75]Braun Michael and McAuliffe Jon, "Variational inference for large-scale models of discrete choice," Journal of the American Statistical Association, vol.105, March 2010.
[76]Kadir Timor and Brady Michael, "Saliency, scale and image description," International Journal of Computer Vision, vol.45,2001, pp.83-105.
[77]Chang Chih-Chung and Lin Chih-Jen, "LIBSVM:A Library for support vector machines," ACM Transactions on Intelligent Systems and Technology, vol.2, 2011.
[78]Mangai M. Alarmel and Gounden N. Ammasai, "Multi-view clustering," in Proceedings-Fourth IEEE International Conference on Data Mining, ICDM 2004, November 1,2004-November 4,2004, Brighton, United kingdom,2004, pp.19-26.
[79]Christoudias C. Mario, Darrell Trevor and Urtasun Raquel., "Multi-view learning in the presence of view disagreement," in 24th Conference on Uncertainty in Artificial Intelligence, July 9,2008-July 12,2008, Helsinki, Finland,2008, pp.88-96.
[80]Zhang Dan, Wang Fei, Zhang Changshui, et al., "Multi-view local learning," in 23rd AAAI Conference on Artificial Intelligence and the 20th Innovative Applications of Artificial Intelligence Conference, July 13,2008-July 17,2008, Chicago, IL, United states,2008, pp.752-757.
[81]Sindhwani Vikas and Rosenberg David S., "An RKHS for multi-view learning and manifold co-regularization," in 25th International Conference on Machine Learning, July 5,2008-July 9,2008, Helsinki, Finland,2008, pp.976-983.
[82]Shan Jiang, Kai Shuang, Fan Guoliang, et al.,"Multi-view face recognition based on manifold learning and multilinear representation," in 2008 9th International Conference on Signal Processing, October 26,2008-October 29,2008, Beijing, China, 2008, pp.2100-2103.
[83]He Jingrui, Lawrence Rick, "A graph-based framework for multi-task multi-view learning," in 28th International Conference on Machine Learning, June 28, 2011-July 2,2011, Bellevue, WA, United states,2011, pp.25-32.
[84]Shen Dan, Zhang Jie, Su Jian, et ai, "A collaborative ability measurement for co-training," in First International Joint Conference on Natural Language Processing, March 22,2004-March 24,2004, Hainan Island, China,2005, pp. 436-445.
[85]Eaton Eric, DesJardins Marie, Jacob Sara, "Multi-view clustering with constraint propagation for learning with an incomplete mapping between views," in 19th International Conference on Information and Knowledge Management and Co-located Workshops, October 26,2010-October 30,2010, Toronto, ON, Canada, 2010, pp.389-398.
[86]Muslea Ion, Minton Steven, Knoblock Craig A., "Active+ Semi-Supervised Learning=Robust Multi-View Learning," presented at the The Nineteenth International Conference on Machine Learning (ICML2002), The University of New South Wales Sydney Australia 2002.
[87]Wang Wei and Zhou Zhi-Hua, "A new analysis of co-training," in 27th International Conference on Machine Learning, June 21,2010-June 25,2010, Haifa, Israel,2010, pp.1135-1142.
[88]Yaslan Yusuf and Cataltepe Zehra, "Co-training with adaptive Bayesian classifier combination," in 2008 23rd International Symposium on Computer and Information Sciences, October 27,2008-October 29,2008, Istanbul, Turkey,2008.
[89]Yu Shipeng, Krishnapuram Balaji, Rosales Romer, et al.,"Bayesian co-training," Journal of Machine Learning Research, vol.12,2011, pp.2649-2680.
[90]Cawley Gavin C., Talbot Nicola L. C. and Girolami Mark, "Sparse Multinomial Logistic Regression via Bayesian L1 Regularisation," presented at the Neural Information Processing Systems Conference (NIPs), British Columbia, Canada 2007.
[91]Doktorski L., "L2-SVM:Dependence on the regularization parameter," Pattern Recognition and Image Analysis, vol.21, pp.254-257,2011.
[92]Cortes Corinna, Mohri Mehryar, Rostamizadeh Afshin,"L2 regularization for learning kernels," in 25th Conference on Uncertainty in Artificial Intelligence,June 18,2009-June 21,2009, Montreal, QC, Canada,2009, pp.109-116.
[93]Mazilu Sinziana and Iria Jose, "L1 vs. L2 regularization in text classification when learning from labeled features," in 10th International Conference on Machine Learning and Applications, December 18,2011-December 21,2011, Honolulu, HI, United states,2011, pp.166-171.
[94]Krishnapuram Balaji, Carin Lawrence, Figueiredo Mario A.T., et al., "Sparse multinomial logistic regression:Fast algorithms and generalization bounds," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.27,2005, pp. 957-968.
[95]Laplace distribution. Available: http://en.wikipedia.org/wiki/Laplace_distribution
[96]Yuan Guo-Xun, Ho Chia-Hua and Lin Chih-Jen, "An improved glmnet for L1-regularized logistic regression," in 17 th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, August 21,2011-August 24,2011, San Diego, CA, United states,2011, pp.33-41.
[97]Cen Ling, Yu Zhu Liang, Dong Ming Hui, "Speech emotion recognition system based on L1 regularized linear regression and decision fusion," in 4th International Conference on Affective Computing and Intelligent Interaction, October 9, 2011-October 12,2011, Memphis, TN, United states,2011, pp.332-340.
[98]Wang Feichao, "A survey on automatic image annotation and trends of the new age," in 2011 International Conference on Power Electronics and Engineering Application, December 24,2011-December 25,2011, Shenzhen, China,2011, pp. 434-438.
[99]Wang Changhu, Jing Feng, Zhang Lei, et al, "Scalable search-based image annotation," Multimedia Systems, vol.14, pp.205-220,2008.
[100]Ke Xiao, Li Shaozi, CaoDonglin, "A two-level model for automatic image annotation," Multimedia Tools and Applications,2011, pp.1-18.
[101]Kang Feng, Jin Rong, Chai Joyce Y., "Regularizing translation models for better automatic image annotation," in CIKM 2004:Proceedings of the Thirteenth ACM Conference on Information and Knowledge Management, November 8,2004 November 13,2004, Washington, DC, United states,2004, pp.350-359.
[102]Guo Yu Tang, Luo Bin, "An automatic image annotation method based on the mutual K-nearest neighbor graph," in 2010 6th International Conference on Natural Computation, August 10,2010-August 12,2010, Yantai, Shandong, China,2010, pp. 3562-3566.
[103]Kanemura Atsunori, Maeda Shin-ichi and Ishii Shin, "Superresolution with compound Markov random fields via the variational EM algorithm," Neural Networks, vol.22,2009, pp.1025-1034.
[104]Huang Hao, Bi Le-Peng, Song Han-Tao and Lu Yu-Chang., "A variational EM algorithm for large databases," in International Conference on Machine Learning and Cybernetics, August 18,2005-August 21,2005, Guangzhou, China,2005, pp. 3048-3052.
[105]Liu Qiang and Ihler Alexander, "Variational algorithms for marginal MAP," in 27th Conference on Uncertainty in Artificial Intelligence, July 14,2011-July 17,2011, Barcelona, Spain,2011, pp.453-462.
[106]Takasu Atsuhiro, Fukagawa Daiji and Akutsu Tatsuya, "A variational Bayesian EM algorithm for tree similarity," in 2010 20th International Conference on Pattern Recognition, August 23,2010-August 26,2010, Istanbul, Turkey,2010, pp. 1056-1059.
[107]Li Xiaoxu, Wang Xiaojie, Wu Chunxiao, et al., "Correspondence with category Latent Dirichlet Allocation for image annotation," in 2nd International Conference on Multimedia Technology, July 26,2011-July 28,2011, Hangzhou, China, 2011, pp.4879-4882.
[108]Torralba Antonio and Efros Alexei A., "Unbiased look at dataset bias," in 2011 IEEE Conference on Computer Vision and Pattern Recognition, June 20,2011 June 25,2011, Colorado Springs, CO, United states,2011, pp.1521-1528.
[109]钟义信,信息科学原理.北京：北京邮电大学出版社,2002.
[2]钟义信,机器知行学原理.北京：北京出版社,2007.
[3]Bishop Christopher M., Pattern Recognition and Machine Learning second ed. Secaucus, NJ, USA:Springer-Verlag New York,2006.
[4]Tarek Habib, Jordi Inglada, Gregoire Mercier, et al., "Speeding up support vector machine (SVM) image classification by a kernel series expansion," in 2008 IEEE International Conference on Image Processing, October 12,2008-October 15,2008, San Diego, CA, United states,2008, pp.865-868.
[5]Begum Demir and Erturk Sarp, "Improving SVM classification accuracy using a hierarchical approach for hyperspectral images," in 2009 IEEE International Conference on Image Processing, November 7,2009-November 12,2009, Cairo, Egypt, 2009, pp.2849-2852.
[6]沈新宇,许宏丽,官腾飞,”基于直推式支持向量机的图像分类算法,”计算机应用,vol.6,2007.
[7]Bosch Anna, Zierssman Andrew and Munoz, Xavierl, "Image classification using random forests and ferns," in 2007 IEEE 11th International Conference on Computer Vision, October 14,2007-October 21,2007, Rio de Janeiro, Brazil,2007.
[8]Li, Yan and Li, Lei, "A novel split and merge EM algorithm for Gaussian mixture model," in 5th International Conference on Natural Computation, August 14, 2009-August 16,2009, Tianjian, China,2009, pp.479-483.
[9]Vailaya Aditya, Figueiredo Mario A. T., JainAnil K., et al. "Image classification for content-based indexing," IEEE Transactions on Image Processing, vol. 10,2001, pp.117-130.
[10]Hofmann Thomas, "Unsupervised learning by probabilistic Latent Semantic Analysis," Machine Learning, vol.42,2001, pp.177-196.
[11]Blei David M., Ng Andrew Y. and Jordan Michael I., "Latent Dirichlet allocation," Journal of Machine Learning Research, vol.3,2003, pp.993-1022.
[12]Sivic Josef, Russell Bryan C.,Efros Alexei A., et al, "Discovering objects and their location in images," in Proceedings-10th IEEE International Conference on Computer Vision, October 17,2005-October 20,2005, Beijing, China, 2005, pp.370-377.
[13]Quelhas P., Monay F., Odobez, J.-M. Gatica-Perez D., et al., "Modeling scenes with local descriptors and latent aspects," in Proceedings-10th IEEE International Conference on Computer Vision, October 17,2005-October 20,2005, Beijing, China,2005, pp.883-890.
[14]Bosch Anna, Zisserman Andrew, Munoz Xavier, "Scene classification via pLSA," in 9th European Conference on Computer Vision, May 7,2006-May 13,2006, Graz, Austria,2006, pp.517-530.
[15]Li Fei-Fei and Perona Pietro, "A bayesian hierarchical model for learning natural scene categories," in 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 20,2005-June 25,2005, San Diego, CA, United states,2005, pp.524-531.
[16]Wang Chong, Blei David and Li Fei-Fei, "Simultaneous image classification and annotation," in 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, June 20,2009-June 25,2009, Miami, FL, United states,2009, pp.1903-1910.
[17]Yang Yi, Wu Fei, Nie Feiping, et al., "Web and personal image annotation by mining label correlation with relaxed visual graph embedding," IEEE Transactions on Image Processing, vol.21,2012, pp.1339-1351.
[18]Li Xirong, Gavves Efstratios, Snoek Cees G.M., et al., "Personalizing automated image annotation using cross-entropy," in 19th ACM International Conference on Multimedia ACM Multimedia 2011, November 28,2011-December 1, 2011, Scottsdale, AZ, United states,2011, pp.233-242.
[19]Wang Changhu, Jing Feng, Zhang Lei, et al., "Content-based image annotation refinement," in 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 17,2007-June 22,2007, Minneapolis, MN, United states,2007.
[20]Karadag Ozge Oztimur and Vural Fatos T. Yarman, "HANOLISTIC:A hierarchical automatic image annotation system using holistic approach," in 2009 IEEE Conference on Computer Vision and Pattern Recognition, June 20,2009-June 25, 2009, Miami, FL, United states,2009, pp.16-21.
[21]Zhao Yufeng, Zhao Yao, Zhu Zhenfeng, "Pseudo relevance feedback for searech-based image annotation," Chinese Journal of Electronics, vol.17,2008, pp. 471-474.
[22]Wang Hua, Huang Heng, Ding Chris, "Image annotation using multi-label correlated Green's function," in 12th International Conference on Computer Vision, September 29,2009-October 2,2009, Kyoto, Japan,2009, pp.2029-2034.
[23]Mueen A.l,Zainuddin R., Baba M. Sapiyan, "Automatic multilevel medical image annotation and retrieval," Journal of Digital Imaging, vol.21,2008, pp. 290-295.
[24]Barnard K. and Forsyth D., "Learning the semantics of words and pictures," in 8th International Conference on Computer Vision, July 9,2001-July 12, 2001, Vancouver, BC, United states,2001, pp.408-415.
[25]Gao Yan-Yu, Yin Yi-Xin, Uozumi Takashi, "A hierarchical image annotation method based on SVM and semi-supervised EM," Zidonghua Xuebao/Acta Automatica Sinica, vol.36,2010, pp.960-967.
[26]Qi Xiaojun and Han Yutao, "Incorporating multiple SVMs for automatic image annotation," Pattern Recognition, vol.40,2007, pp.728-741.
[27]Han Yutao, Qi Xiaojun, "A complementary SVMs-based image annotation system," in IEEE International Conference on Image Processing 2005, ICIP 2005, September 11,2005-September 14,2005, Genova, Italy,2005, pp.1185-1188.
[28]Zhao Yufeng, Zhao Yao, Zhu Zhenfeng, "A novel image annotation scheme based on neural network," in 8th International Conference on Intelligent Systems Design and Applications, November 26,2008-November 28,2008, Kaohsiung, Taiwan, 2008, pp.644-647.
[29]王科平,”自动图像标注的关键技术研究,”博士论文,北京邮电大学,2011.
[30]Feng Songhe, Xu De, Li Bing, "Automatic region-based image annotation using an improved multiple-instance learning algorithm," Chinese Journal of Electronics, vol.17,2008,pp.43-47.
[31]Yang Changbo, Dong Ming and Hua Jing, "Region-based image annotation using asymmetrical support vector machine-based multiple-instance learning," in 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 17,2006-June 22,2006, New York, NY, United states,2006, pp.2057-2063.
[32]Putthividhya D., Attias H.T. and Nagarajan, S.S., "Supervised topic model for automatic image annotation," in 2010 IEEE International Conference on Acoustics Speech, and Signal Processing, March 14,2010-March 19,2010, Dallas, TX, United states,2010, pp.1894-1897.
[33]Blei David M. and McAuliffe Jon D., "Supervised topic models," in 21st Annual Conference on Neural Information Processing Systems, December 3,2007 December 6,2007, Vancouver, BC, Canada,2009.
[34]Blei David M. and Jordan Michael I., "Modeling Annotated Data," in Proceedings of the Twenty-Sixth Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, July 28,2003-August 1,2003, Toronto, Ont., Canada,2003, pp.127-134.
[35]Jeon J., Lavrenko V. and Manmatha R., "Automatic Image Annotation and Retrieval using Cross-Media Relevance Models," in Proceedings of the Twenty-Sixth Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, July 28,2003-August 1,2003, Toronto, Ont., Canada,2003, pp. 119-126.
[36]Lavrenko V, Manmatha and R. J. Jeon, "A model for leaning the semantics of pictures," presented at the Neural Information Processing Systems (NIPS 2004).
[37]Putthividhya Duangmanee, Attias Hagai T. and Nagarajan Srikantan S., "Topic regression multi-modal Latent Dirichlet Allocation for image annotation," in 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 13,2010-June 18,2010, San Francisco, CA, United states,2010, pp. 3408-3415.
[38]bag of words. Available: http://en.wikipedia.org/wiki/Bag_of_words_model
[39]Zhang Yin, Jin Rong and Zhou Zhi-Hua, "Understanding bag-of-words model: A statistical framework," International Journal of Machine Learning and Cybernetics, vol.1,2010, pp.43-52.
[40]Wu Lei, Hoi, Steven C. H. and Yu Nenghai, "Semantics-preserving bag-of-words models and applications," IEEE Transactions on Image Processing, vol. 19,2010, pp.1908-1920.
[41]Li Teng, Mei Tao, Kweon In-So, Hua Xian-Sheng, "Contextual bag-of-words for visual categorization," IEEE Transactions on Circuits and Systems for Video Technology, vol.21, pp.381-392,2011.
[42]Lowe David G.., "Object recognition from local scale-invariant features," in Proceedings of the 1999 7th IEEE International Conference on Computer Vision, September 20,1999-September 27,1999, Kerkyra, Greece,1999, pp.1150-1157.
[43]Wu Jianxin, Rehg Jim M., "CENTRIST:A visual descriptor for scene categorization," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 33,2011,pp.1489-1501.
[44]Wu Jianxin and Rehg James M., "Beyond the Euclidean distance:Creating effective visual codebooks using the histogram intersection kernel," in 12th International Conference on Computer Vision, September 29,2009-October 2,2009, Kyoto, Japan,2009, pp.630-637.
[45]Bayesian network. Available: http://en.wikipedia.org/wiki/Bayesian_network
[46]Liang Lin, Wen Fang, Xu Ying-Qing, et al., "Accurate face alignment using shape constrained Markov network," in 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 17,2006-June 22,2006, New York, NY, United states,2006, pp.1313-1320.
[47]Barrat Sabine and Tabbone Salvatore, "Modeling, classifying and annotating weakly annotated images using Bayesian network," Journal of Visual Communication and Image Representation, vol.21,2010, pp.355-363.
[48]Alam Fahim Irfanl and Gondra Iker, "A Bayesian network-based tunable image segmentation algorithm for object recognition," in 11th IEEE International Symposium on Signal Processing and Information Technology, December 14,2011 December 17,2011, Bilbao, Spain,2011, pp.011-016.
[49]Dewar Michael, Wiggins Chris and Wood Frank, "Inference in hidden Markov models with explicit state duration distributions," IEEE Signal Processing Letters, vol.19,2012, pp.235-238.
[50]Markov Random Field. Available: http://en.wikipedia.org/wiki/Markoy_random_field
[51]Galecki Andrzej T., Ten Have Thomas R. and Molenberghs Geert, "Simple and fast alternative to the EM algorithm for incomplete categorical data and latent class models," Computational Statistics and Data Analysis, vol.35,2001, pp.265-281.
[52]Teh Yee Whye, Kurihara Kenichi and Welling Max, "Collapsed variational inference for HDP," in 21st Annual Conference on Neural Information Processing Systems, December 3,2007-December 6,2007, Vancouver, BC, Canada,2009.
[53]Wang Chong and Blei David M., "Variational inference for the nested Chinese restaurant process," in 23rd Annual Conference on Neural Information Processing Systems, December 7,2009-December 10,2009, Vancouver, BC, Canada, 2009, pp.1990-1998.
[54]Liao Chih-Pin and Chien Jen-Tzung, "Variational inference for conditional random fields," in 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, March 14,2010-March 19,2010, Dallas, TX, United states,2010, pp.2002-2005.
[55]Monte Carlo. Available:http://en.wikipedia.org/wiki/Monte_Carlo_method
[56]Jordan Michael I., "An introduction to variational methods for graphical models," Machine Learning, vol.37, pp.183-233,1999.
[57]maximum entropy. Available: http://en.wikipedia.org/wiki/Maximum_entropy
[58]Kullback-Leibler divergence. Available: http://en.wikipedia.org/wiki/Kullback%E2%80%93Leibler_divergence
[59]Tanaka Toshiyuki, "Information geometry of mean field theory," IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, 1996, vol. E79-A, pp.709-715.
[60]mean field theory. Available: http://en.wikipedia.org/wiki/Mean_field_theory
[61]Generative Model. Available: http://en.wikipedia.org/wiki/Generative_model
[62]Dirichlet Distribution. Available: http://en.wikipedia.org/wiki/Pirichlet_distribution
[63]Multinomial Distribution. Available: http://en.wikipedia.org/wiki/Multinomial_distribution
[64]Simplex. Available:http://en.wikipedia.org/wiki/Simplex
[65]Gamma function. Available:http://en.wikipedia.org/wiki/Gamma_function
[66]Nocedal Jorge and Wright Stephen J., Numerical Optimization. Berlin, Germany:Springer,2006.
[67]Russell Bryan C., Torralba Antonio, Murphy Kevin P., et al., "LabelMe:A database and web-based tool for image annotation," International Journal of Computer Vision, vol.77,2008, pp.157-173.
[68]Cao Liangliang, Li Fei-Fei, "Spatially coherent latent topic model for concurrent segmentation and classification of objects and scenes," in 2007 IEEE 11th International Conference on Computer Vision, October 14,2007-October 21,2007, Rio de Janeiro, Brazil,2007.
[69]Nishii Ryuei and Eguchi Shinto, "Supervised image classification based on AdaBoost with contextual weak classifiers," in 2004 IEEE International Geoscience and Remote Sensing Symposium Proceedings:Science for Society:Exploring and Managing a Changing Planet. September 20,2004-September 24,2004, Anchorage, AK, United states,2004, pp.1467-1470.
[70]Gao Feng and Ai Haizhou, "Face age classification on consumer images with gabor feature and fuzzy LDA method," in 3rd International Conference on Advances in Biometrics, June 2,2009-June 5,2009, Alghero, Italy,2009, pp.132-141.
[71]Chen Kai and Jin Yang, "An ensemble learning algorithm based on Lasso selection," in 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, October 29,2010-October 31,2010, Xiamen, China,2010, pp. 617-620.
[72]Liuxia and Lisu, "Research on Face Detection Classifier using an improved Adboost algorithm," in International Symposium on Computer Science and Computational Technology, December 20,2008-December 22,2003, Shanghai, China, 2008, pp.78-81.
[73]Wang Gang, Hoiem Derek and Forsyth David, "Building text features for object image classification," in 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, June 20,2009-June 25,2009, Miami, FL, United states,2009, pp.1367-1374.
[74]Li Xiaoxu, Zeng Junyu,Wang Xiaojie, et al., "Mixture of softmax sLDA," in 11th IEEE International Conference on Data Mining, ICDM 2011, December 11, 2011-December 14,2011, Vancouver, BC, Canada,2011, pp.1164-1169.
[75]Braun Michael and McAuliffe Jon, "Variational inference for large-scale models of discrete choice," Journal of the American Statistical Association, vol.105, March 2010.
[76]Kadir Timor and Brady Michael, "Saliency, scale and image description," International Journal of Computer Vision, vol.45,2001, pp.83-105.
[77]Chang Chih-Chung and Lin Chih-Jen, "LIBSVM:A Library for support vector machines," ACM Transactions on Intelligent Systems and Technology, vol.2, 2011.
[78]Mangai M. Alarmel and Gounden N. Ammasai, "Multi-view clustering," in Proceedings-Fourth IEEE International Conference on Data Mining, ICDM 2004, November 1,2004-November 4,2004, Brighton, United kingdom,2004, pp.19-26.
[79]Christoudias C. Mario, Darrell Trevor and Urtasun Raquel., "Multi-view learning in the presence of view disagreement," in 24th Conference on Uncertainty in Artificial Intelligence, July 9,2008-July 12,2008, Helsinki, Finland,2008, pp.88-96.
[80]Zhang Dan, Wang Fei, Zhang Changshui, et al., "Multi-view local learning," in 23rd AAAI Conference on Artificial Intelligence and the 20th Innovative Applications of Artificial Intelligence Conference, July 13,2008-July 17,2008, Chicago, IL, United states,2008, pp.752-757.
[81]Sindhwani Vikas and Rosenberg David S., "An RKHS for multi-view learning and manifold co-regularization," in 25th International Conference on Machine Learning, July 5,2008-July 9,2008, Helsinki, Finland,2008, pp.976-983.
[82]Shan Jiang, Kai Shuang, Fan Guoliang, et al.,"Multi-view face recognition based on manifold learning and multilinear representation," in 2008 9th International Conference on Signal Processing, October 26,2008-October 29,2008, Beijing, China, 2008, pp.2100-2103.
[83]He Jingrui, Lawrence Rick, "A graph-based framework for multi-task multi-view learning," in 28th International Conference on Machine Learning, June 28, 2011-July 2,2011, Bellevue, WA, United states,2011, pp.25-32.
[84]Shen Dan, Zhang Jie, Su Jian, et ai, "A collaborative ability measurement for co-training," in First International Joint Conference on Natural Language Processing, March 22,2004-March 24,2004, Hainan Island, China,2005, pp. 436-445.
[85]Eaton Eric, DesJardins Marie, Jacob Sara, "Multi-view clustering with constraint propagation for learning with an incomplete mapping between views," in 19th International Conference on Information and Knowledge Management and Co-located Workshops, October 26,2010-October 30,2010, Toronto, ON, Canada, 2010, pp.389-398.
[86]Muslea Ion, Minton Steven, Knoblock Craig A., "Active+ Semi-Supervised Learning=Robust Multi-View Learning," presented at the The Nineteenth International Conference on Machine Learning (ICML2002), The University of New South Wales Sydney Australia 2002.
[87]Wang Wei and Zhou Zhi-Hua, "A new analysis of co-training," in 27th International Conference on Machine Learning, June 21,2010-June 25,2010, Haifa, Israel,2010, pp.1135-1142.
[88]Yaslan Yusuf and Cataltepe Zehra, "Co-training with adaptive Bayesian classifier combination," in 2008 23rd International Symposium on Computer and Information Sciences, October 27,2008-October 29,2008, Istanbul, Turkey,2008.
[89]Yu Shipeng, Krishnapuram Balaji, Rosales Romer, et al.,"Bayesian co-training," Journal of Machine Learning Research, vol.12,2011, pp.2649-2680.
[90]Cawley Gavin C., Talbot Nicola L. C. and Girolami Mark, "Sparse Multinomial Logistic Regression via Bayesian L1 Regularisation," presented at the Neural Information Processing Systems Conference (NIPs), British Columbia, Canada 2007.
[91]Doktorski L., "L2-SVM:Dependence on the regularization parameter," Pattern Recognition and Image Analysis, vol.21, pp.254-257,2011.
[92]Cortes Corinna, Mohri Mehryar, Rostamizadeh Afshin,"L2 regularization for learning kernels," in 25th Conference on Uncertainty in Artificial Intelligence,June 18,2009-June 21,2009, Montreal, QC, Canada,2009, pp.109-116.
[93]Mazilu Sinziana and Iria Jose, "L1 vs. L2 regularization in text classification when learning from labeled features," in 10th International Conference on Machine Learning and Applications, December 18,2011-December 21,2011, Honolulu, HI, United states,2011, pp.166-171.
[94]Krishnapuram Balaji, Carin Lawrence, Figueiredo Mario A.T., et al., "Sparse multinomial logistic regression:Fast algorithms and generalization bounds," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.27,2005, pp. 957-968.
[95]Laplace distribution. Available: http://en.wikipedia.org/wiki/Laplace_distribution
[96]Yuan Guo-Xun, Ho Chia-Hua and Lin Chih-Jen, "An improved glmnet for L1-regularized logistic regression," in 17 th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, August 21,2011-August 24,2011, San Diego, CA, United states,2011, pp.33-41.
[97]Cen Ling, Yu Zhu Liang, Dong Ming Hui, "Speech emotion recognition system based on L1 regularized linear regression and decision fusion," in 4th International Conference on Affective Computing and Intelligent Interaction, October 9, 2011-October 12,2011, Memphis, TN, United states,2011, pp.332-340.
[98]Wang Feichao, "A survey on automatic image annotation and trends of the new age," in 2011 International Conference on Power Electronics and Engineering Application, December 24,2011-December 25,2011, Shenzhen, China,2011, pp. 434-438.
[99]Wang Changhu, Jing Feng, Zhang Lei, et al, "Scalable search-based image annotation," Multimedia Systems, vol.14, pp.205-220,2008.
[100]Ke Xiao, Li Shaozi, CaoDonglin, "A two-level model for automatic image annotation," Multimedia Tools and Applications,2011, pp.1-18.
[101]Kang Feng, Jin Rong, Chai Joyce Y., "Regularizing translation models for better automatic image annotation," in CIKM 2004:Proceedings of the Thirteenth ACM Conference on Information and Knowledge Management, November 8,2004 November 13,2004, Washington, DC, United states,2004, pp.350-359.
[102]Guo Yu Tang, Luo Bin, "An automatic image annotation method based on the mutual K-nearest neighbor graph," in 2010 6th International Conference on Natural Computation, August 10,2010-August 12,2010, Yantai, Shandong, China,2010, pp. 3562-3566.
[103]Kanemura Atsunori, Maeda Shin-ichi and Ishii Shin, "Superresolution with compound Markov random fields via the variational EM algorithm," Neural Networks, vol.22,2009, pp.1025-1034.
[104]Huang Hao, Bi Le-Peng, Song Han-Tao and Lu Yu-Chang., "A variational EM algorithm for large databases," in International Conference on Machine Learning and Cybernetics, August 18,2005-August 21,2005, Guangzhou, China,2005, pp. 3048-3052.
[105]Liu Qiang and Ihler Alexander, "Variational algorithms for marginal MAP," in 27th Conference on Uncertainty in Artificial Intelligence, July 14,2011-July 17,2011, Barcelona, Spain,2011, pp.453-462.
[106]Takasu Atsuhiro, Fukagawa Daiji and Akutsu Tatsuya, "A variational Bayesian EM algorithm for tree similarity," in 2010 20th International Conference on Pattern Recognition, August 23,2010-August 26,2010, Istanbul, Turkey,2010, pp. 1056-1059.
[107]Li Xiaoxu, Wang Xiaojie, Wu Chunxiao, et al., "Correspondence with category Latent Dirichlet Allocation for image annotation," in 2nd International Conference on Multimedia Technology, July 26,2011-July 28,2011, Hangzhou, China, 2011, pp.4879-4882.
[108]Torralba Antonio and Efros Alexei A., "Unbiased look at dataset bias," in 2011 IEEE Conference on Computer Vision and Pattern Recognition, June 20,2011 June 25,2011, Colorado Springs, CO, United states,2011, pp.1521-1528.
[109]钟义信,信息科学原理.北京：北京邮电大学出版社,2002.