Abstract
As the feature dimension increases, the original pyramid matching kernel suffers from distortion factors that increase linearly with the feature dimension. This paper proposes a new method by consistently dividing the feature space into two subspaces while generating several levels. In each subspace of the level, the original pyramid matching is used. Then, a weighted sum of every subspace at each level is made as the final measurement of similarity. Experiments on data set Caltech-101 and ETH-80 show that compared with other related algorithms which need hundreds of times of original computation time, dimension partition pyramid matching kernel only needs about 4–6 times less of original computation time to obtain the similar accuracy.
Similar content being viewed by others
References
Grauman K, Darrell T (2005) The pyramid match kernel: discriminative classification with sets of image features. In: Proceedings of the IEEE conference on ICCV, Beijing. IEEE, pp 1458–1465
Grauman K, Darrell T (2007) Approximate correspondences in high dimensions. Adv Neural Inf Process Syst 19:505
Lazebnik S, Schmid C, Ponce J (2006) Beyond bags of features: spatial pyramid matching for recognizing natural scene categories. In: Proceedings of the IEEE conference on CVPR. IEEE, New York, pp 2169–2178
Liu Y, Wang X, Zha H (2008) Dimension amnesic pyramid match kernel. In: Proceedings of the 23rd national conference on artificial intelligence, Chicago. AAAI Press, pp 652–658
Zhang J, Zhao G, Gu H (2010) An improved pyramid matching kernel. In: 2010 International conference on life system modeling and simulation & 2010 international conference on intelligent computing for sustainable energy and environment (LSMS & ICSEE 2010), vol 6328, pp 52–61
Grauman K, Darrell T (2004) Fast contour matching using approximate earth mover’s distance. In: Proceedings of the 2004 IEEE computer society conference on computer vision and pattern recognition, Washington. IEEE, pp 220–227
Berg A, Berg T, Malik J (2005) Shape matching and object recognition using low distortion correspondences. In: Proceedings of the IEEE conference on CVPR, San Diego. IEEE, pp 26–33
Lyu S (2005) Mercer kernels for object recognition with local features. In: Proceedings of the IEEE conference on CVPR, San Diego. IEEE, pp 223–229
Charikar M (2002) Similarity estimation techniques from rounding algorithms. In: Proceedings of the 34th annual ACM symposium on theory of computing, Montreal. Association for Computing Machinery, pp 380–388
Indyk P, Thaper N (2003) Fast image retrieval via embeddings. In: 3rd International workshop on statistical and computational theories of vision, Nice
Agarwal P, Varadarajan K (2004) A near-linear algorithm for Euclidean bipartite matching. In: Symposium on computational geometry
Grauman K, Darrell T (2007) The pyramid match kernel: efficient learning with sets of features. J Mach Learn Res 8:725–760
Grauman K (2006) Matching sets of features for efficient retrieval and recognition. MIT, Cambridge
Kapoor A, Shenoy P, Tan D (2008) Combining brain computer interfaces with vision for object categorization. In: Proceedings of the IEEE conference on CVPR, Alaska. IEEE, pp 1–8
Saenko K, Darrell T (2009) Filtering abstract senses from image search results. In: Proceedings of the 23rd annual conference on NIPS, Vancouver. MIT Press, pp 1–9
Bosch A, Zisserman A, Munoz X (2007) Representing shape with a spatial pyramid kernel. In: Proceedings of the 6th ACM international conference on image and video retrieval, Amsterdam. Association for Computing Machinery, pp 401–408
Yeh T, Lee J, Darrell T, MIT C (2007) Adaptive vocabulary forests for dynamic indexing and category learning. In: Proceedings of the IEEE international conference on computer vision, Rio de Janeiro. IEEE, pp 1–8
Fu S, Shengyang G, Hou Z et al (2008) Multiple kernel learning from sets of partially matching image features. In: Proceedings of international conference on pattern recognition, Florida. IEEE Computer Society, pp 1–4
He J, Chang S, Xie L (2008) Fast kernel learning for spatial pyramid matching. In: Proceedings of the IEEE conference on computer vision and pattern recognition, Alaska. IEEE, pp 1–7
Shawe-Taylor J, Cristianini N (2004) Kernel methods for pattern analysis. Cambridge University Press, Cambridge
Lee JJ (2008, Apr) LIBPMK: a pyramid match toolkit. Available from: http://hdl.handle.net/1721.1/41070
Fei-Fei L, Fergus R, Perona P (2004) Learning generative visual models from few training examples. In: Workshop on generative model based vision, Washington. Academic Press, p 178
Harris C, Stephens M (1988) A combined corner and edge detector. In: Proceedings of the 4th Alvey vision conference, Manchester, pp 147–151
Lowe DG (2004) Distinctive image features from scale-invariant keypoints. Int J Comput Vis 60(2):91–110
Leibe B, Schiele B (2003) Analyzing appearance and contour based methods for object categorization. In: Proceedings of the IEEE conference on CVPR, Madison. IEEE, pp 409–415
Ke Y, Sukthankar R (2004) PCA-SIFT: a more distinctive representation for local image descriptors. In: Proceedings of the IEEE computer society conference on computer vision and pattern recognition, Washington. IEEE, pp 503–516
Acknowledgments
This work is supported by National Natural Science Foundation of China (60872070); Zhejiang Province Key Scientific and Technological Project (Grant No. 2007C11094, No. 2008C21141); Zhejiang Provincial Natural Science Foundation of China (Grant No. Y1080766).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Zhang, J., Zhao, G. & Gu, H. DP-PMK: an improved pyramid matching kernel for approximating correspondences in high dimensions. Neural Comput & Applic 21, 1167–1175 (2012). https://doi.org/10.1007/s00521-012-0953-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-012-0953-y