Generative models for similarity-based classification

Cazzanti, Luca; Gupta, Manisha; Koppal, A.J.

doi:10.1016/j.patcog.2008.01.005

Cited by 17 publications

(19 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In similarity based classification (see, e.g., Cazzanti et al 2008;), an index is defined to measure similarity of an observation with respect to training sample observations, and those similarity measures are used as features to develop a classifier. For instance, a nonlinear support vector machine (SVM) (see, e.g., Vapnik 1998) can be viewed as a similarity based classifier, where a kernel function is used to measure similarity/dissimilarity between two observations.…”

Section: Theorem 2(b) Suppose That the J Competing Classes Satisfy Asmentioning

confidence: 99%

“…Graepel et al (1999) and Pekalska et al (2001) used several standard learning techniques on these similarity based features. A discussion on similarity based classifiers including SVM, kernel Fisher discriminant analysis and those based on entropy can be found in Cazzanti et al (2008). It has been observed in the literature that NN classifiers on similarity measures usually yield low misclassification rates (see, e.g., Cost and Salzberg 1993;Pekalska et al 2001).…”

Section: Theorem 2(b) Suppose That the J Competing Classes Satisfy Asmentioning

confidence: 99%

See 1 more Smart Citation

On some transformations of high dimension, low sample size data for nearest neighbor classification

Dutta

Ghosh

2015

Mach Learn

View full text Add to dashboard Cite

For data with more variables than the sample size, phenomena like concentration of pairwise distances, violation of cluster assumptions and presence of hubness often have adverse effects on the performance of the classic nearest neighbor classifier. To cope with such problems, some dimension reduction techniques like those based on random linear projections and principal component directions have been proposed in the literature. In this article, we construct nonlinear transformations of the data based on inter-point distances, which also lead to reduction in data dimension. More importantly, for such high dimension low sample size data, they enhance separability among the competing classes in the transformed space. When the classic nearest neighbor classifier is used on the transformed data, it usually yields lower misclassification rates. Under appropriate regularity conditions, we derive asymptotic results on misclassification probabilities of nearest neighbor classifiers based on the l 2 norm and the l p norms (with p ∈ (0, 1]) in the transformed space, when the training sample size remains fixed and the dimension of the data grows to infinity. Strength of the proposed transformations in the classification context is demonstrated by analyzing several simulated and benchmark data sets.

show abstract

Section: Theorem 2(b) Suppose That the J Competing Classes Satisfy Asmentioning

confidence: 99%

Section: Theorem 2(b) Suppose That the J Competing Classes Satisfy Asmentioning

confidence: 99%

On some transformations of high dimension, low sample size data for nearest neighbor classification

Dutta

Ghosh

2015

Mach Learn

View full text Add to dashboard Cite

show abstract

“…To do this, classifier parameters and/or knowledge bases are updated with fresh incoming data over the time. Since the similarity-based classifiers are actually most commonly used classifier [5], an adaptive form of them is utilized in this study.…”

Section: Adaptive Similarity-based Classifiermentioning

confidence: 99%

Just-in-time outdoor color discrimination using adaptive similarity-based classifier

Sojodishijani

Ramli

Rostami

et al. 2010

IEICE Electron. Express

View full text Add to dashboard Cite

Abstract:The color recognition and identification in operation time is a critical task in color-based computer vision applications. The main problem for recognizing the real color arises when the color characteristics are changed dynamically in the life time of a system. The outdoor color models which have been addressed by some researchers have serious practical limitations to employ in real applications. Moreover, due to high fluctuations in environment illumination, using conventional classifier for discriminating colors is a complicated task. In this paper, a just-in-time and model-free solution in order to discriminate outdoor colors on data driven modality is proposed. For this purpose, adaptive similarity-based classifier is utilized to track the color's data evolution during a day. Keywords: Model-free outdoor color constancy, adaptive similaritybased classification, Non-stationarity tracking Classification: Science and engineering for electronics References[1] S. D. Buluswar and B. A. Draper, "Color models for outdoor machine vision,"

show abstract

“…Changes in illumination, viewpoint and pose, occlusions, and background clutter make it difficult to construct a robust appearance model which is capable of adapting to changes in the environment. Many appearance models [1,2,3,4] have been proposed for object tracking. These include models based on histograms, kernel density estimates, Gaussian mixture models (GMMs), conditional random fields, subspaces, and discriminative classification.…”

Section: Introductionmentioning

confidence: 99%

Single and Multiple Object Tracking Using a Multi-Feature Joint Sparse Representation

Wei

Zhang

et al. 2015

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

Abstract:In this paper, we propose a tracking algorithm based on a multi-feature joint sparse representation.The templates for the sparse representation can include pixel values, textures, and edges. In the multi-feature joint optimization, noise or occlusion is dealt with using a set of trivial templates. A sparse weight constraint is introduced to dynamically select the relevant templates from the full set of templates. A variance ratio measure is adopted to adaptively adjust the weights of different features. The multi-feature template set is updated adaptively. We further propose an algorithm for tracking multi-objects with occlusion handling based on the multi-feature joint sparse reconstruction. The observation model based on sparse reconstruction automatically focuses on the visible parts of an occluded object by using the information in the trivial templates. The multi-object tracking is simplified into a joint Bayesian inference. The experimental results show the superiority of our algorithm over several state-of-the-art tracking algorithms.

show abstract

Generative models for similarity-based classification

Cited by 17 publications

References 24 publications

On some transformations of high dimension, low sample size data for nearest neighbor classification

On some transformations of high dimension, low sample size data for nearest neighbor classification

Just-in-time outdoor color discrimination using adaptive similarity-based classifier

Single and Multiple Object Tracking Using a Multi-Feature Joint Sparse Representation

Contact Info

Product

Resources

About