Learning Free Energy Kernel for Image Retrieval = Learning Free Energy Kernel for Image Retrieval

다국어 초록 (Multilingual Abstract)

Content-based image retrieval has been the most important technique for managing huge amount of images. The fundamental yet highly challenging problem in this field is how to measure the content-level similarity based on the low-level image features. The primary difficulties lie in the great variance within images, e.g. background, illumination, viewpoint and pose. Intuitively, an ideal similarity measure should be able to adapt the data distribution, discover and highlight the content-level information, and be robust to those variances. Motivated by these observations, we in this paper propose a probabilistic similarity learning approach. We first model the distribution of low-level image features and derive the free energy kernel (FEK), i.e., similarity measure, based on the distribution. Then, we propose a learning approach for the derived kernel, under the criterion that the kernel outputs high similarity for those images sharing the same class labels and output low similarity for those without the same label. The advantages of the proposed approach, in comparison with previous approaches, are threefold. (1) With the ability inherited from probabilistic models, the similarity measure can well adapt to data distribution. (2) Benefitting from the content-level hidden variables within the probabilistic models, the similarity measure is able to capture content-level cues. (3) It fully exploits class label in the supervised learning procedure. The proposed approach is extensively evaluated on two well-known databases. It achieves highly competitive performance on most experiments, which validates its advantages.

참고문헌 (Reference)

1 L. K. Saul, "Think globally, fit locally: unsupervised learning of low dimensional manifolds" 4 : 119-155, 2003

2 K. Chatfield, "The devil is in the details: an e-valuation of recent feature encoding methods" 2011

3 X. Li, "Stochastic feature mapping for PAC-Bayes classification"

4 A. Webb, "Statistical pattern recognition" Wiley 2003

5 S Zhang, "Semi-supervised distance metric learning for collaborative image retrieval and clustering" 2010

6 P Wu, "Randomly Projected KD-Trees with Distance Metric Learning for Image Retrieval" 371-382, 2011

7 T. Jebara, "Probability product kernels" 5 : 819-844, 2004

8 K. Kim, "Probabilistic cost model for nearest neighbor search in image retrieval" 2012

9 J. Goldberger, "Neighbourhood components analysis" 2004

10 X. He, "Multiple kernel learning via distance metric learning for interactive image retrieval" 6713 : 147-156, 2011

11 A. Oliva, "Modeling the shape of the scene: A holistic representation of the spatial envelope" 42 (42): 145-175, 2001

12 M. Sugiyama, "Local fisher discriminant analysis for supervised dimensionality reduction" 905-912, 2006

13 L. Fei-Fei, "Learning generative visual models from few training examples: An incremental Bayesian approach tested on 101 object categories" 106 (106): 59-70, 2007

14 A. Bar-Hillel, "Learning distance functions using equivalence relations" 20 (20): 2003

15 Laurens van der Maaten, "Learning discriminative fisher kernels" 217-224, 2011

16 M. Belkin, "Laplacian eigenmaps for dimensionality reduction and data representation" 15 (15): 1373-1396, 2003

17 B. Wang, "Integrating distance metric learning into label propagation model for multi-label image annotation" 3649-3652, 2011

18 A. K. Jain, "Image retrieval using color and shape" 29 (29): 1233-1244, 1996

19 X. Li, "Hybrid generative-discriminative classification using posterior divergence" 2713-2720, 2011

20 A. Perina, "Free energy score spaces: Using generative information in discriminative classifiers" 34 (34): 1249-1262, 2012

21 T. Jaakkola, "Exploiting generative models in discriminative classifiers" 487-493, 1999

22 M. Subrahmanyam, "Expert system design using wavelet and color vocabulary trees for image retrieval" 39 (39): 5104-5114, 2012

23 K. Van De Sande, "Evaluating color descriptors for object and scene recognition" 32 (32): 1582-1596, 2010

24 L. Yang, "Distance metric learning: A comprehensive survey" Michigan State Universiy 1-51, 2006

25 E. Xing, "Distance metric learning, with application to clustering with side-information" 15 : 505-512, 2002

26 Y. Ying, "Distance metric learning with eigenvalue optimization" 13 : 1-26, 2012

27 J. Blitzer, "Distance metric learning for large margin nearest neighbor classification" 1473-1480, 2005

28 T. Hastie, "Discriminant adaptive nearest neighbor classification" 18 (18): 607-616, 1996

29 X. Wang, "Contextual weighting for vocabulary tree based image retrieval" 209-216, 2011

30 N. Jhanwar, "Content based image retrieval using motif co-occurrence matrix" 22 (22): 1211-1220, 2004

31 A. D. Holub, "Combining generative models and fisher kernels for object recognition" 1 : 136-143, 2005

32 M. J. Swain, "Color indexing" 7 (7): 11-32, 1991

33 Bin Wang, "Collaborative Similarity Metric Learning for Semantic Image Annotation and Retrieval" 한국인터넷정보학회 7 (7): 1252-1271, 2013

34 X. Li, "Bimodal gender recognition from face and fingerprint" 2X. Zhao 2590-2597, 2010

35 W. Bian, "Biased discriminant Euclidean embedding for content-based image retrieval" 19 (19): 545-554, 2010

36 S. Lazebnik, "Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories" 2 : 2169-2178, 2006

37 J. Puzicha, "Article(CrossRef Link)" 1165-1172, 1999

38 C. -H. Lin, "A smart content-based image retrieval system based on color and texture feature" 27 (27): 658-665, 2009

39 E. Rashedi, "A simultaneous feature adaptation and feature selection method for content-based image retrieval systems" 39 : 85-94, 2013

40 M. E. ElAlami, "A novel image retrieval model based on the most relevant features" 24 (24): 23-32, 2011

41 M. Arevalillo-Herr´aez, "A naive relevance feedback model for content-based image retrieval using multiple similarity measures" 43 (43): 619-629, 2010

42 D. Ziou, "A hybrid probabilistic framework for content-based image retrieval with feature weighting" 42 (42): 1511-1519, 2009

43 J. B. Tenenbaum, "A global geometric framework for non-linear dimensionality reduction" 290 (290): 2319-2323, 2000

44 D. Cerra, "A fast compression-based similarity measure with applications to content-based image retrieval" 23 (23): 293-302, 2012

45 L. Yang, "A boosting framework for visuality-preserving distance metric learning and its application to medical image retrieval" 32 (32): 30-44, 2010