Toward Robust Distance Metric Analysis for Similarity Estimation

In this paper, we present a general guideline to establish the relation between a distribution model and its corresponding similarity estimation. A rich set of distance metrics, such as harmonic distance and geometric distance, is derived according to Maximum Likelihood theory. These metrics can provide a more accurate feature model than the conventional Euclidean distance (SSD) and Manhattan distance (SAD). Because the feature elements are from heterogeneous sources and may have different influence on similarity estimation, the assumption of single isotropic distribution model is often inappropriate. We propose a novel boosted distance metric that not only finds the best distance metric that fits the distribution of the underlying elements but also selects the most important feature elements with respect to similarity. We experiment with different distance metrics for similarity estimation and compute the accuracy of different methods in two applications: stereo matching and motion tracking in video sequences. The boosted distance metric is tested on fifteen benchmark data sets from the UCI repository and two image retrieval applications. In all the experiments, robust results are obtained based on the proposed methods.

[1]  B. Ripley,et al.  Robust Statistics , 2018, Encyclopedia of Mathematical Geosciences.

[2]  Yoram Singer,et al.  Improved Boosting Algorithms Using Confidence-rated Predictions , 1998, COLT' 98.

[3]  Dimitrios Gunopulos,et al.  Locally Adaptive Metric Nearest-Neighbor Classification , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Nicu Sebe,et al.  Boosting the distance estimation: Application to the K-Nearest Neighbor Classifier , 2006, Pattern Recognit. Lett..

[5]  Linda G. Shapiro,et al.  Computer and Robot Vision , 1991 .

[6]  Nicu Sebe,et al.  Toward Improved Ranking Metrics , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Hong Yan,et al.  Comparison of face verification results on the XM2VTFS database , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[8]  Aleix M. Martinez,et al.  The AR face database , 1998 .

[9]  Michael J. Swain,et al.  Color indexing , 1991, International Journal of Computer Vision.

[10]  Tomer Hertz,et al.  Learning Distance Functions using Equivalence Relations , 2003, ICML.

[11]  Peter J. Huber,et al.  Robust Statistics , 2005, Wiley Series in Probability and Statistics.

[12]  Jing Peng,et al.  LDA/SVM driven nearest neighbor classification , 2003, IEEE Trans. Neural Networks.

[13]  Tomer Hertz,et al.  Learning distance functions for image retrieval , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[14]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[15]  Michael I. Jordan,et al.  Distance Metric Learning with Application to Clustering with Side-Information , 2002, NIPS.

[16]  Mohan S. Kankanhalli,et al.  Shape Measures for Content Based Image Retrieval: A Comparison , 1997, Inf. Process. Manag..

[17]  Moshe Zakai General error criteria (Corresp.) , 1964, IEEE Trans. Inf. Theory.

[18]  A. Martínez,et al.  The AR face databasae , 1998 .