Deriving TF-IDF as a Fisher Kernel

Elkan, Charles

doi:10.1007/11575832_33

Cited by 28 publications

(13 citation statements)

References 5 publications

(5 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the field of text processing, [8] connected the Fisher kernel and tf-idf scoring in the case of a specific model: the Dirichlet compound multinomial. We now show that the Fisher vector as applied to a GMM extends the tf-idf scoring to continuous events, where an event is the occurrence of a descriptor in an image.…”

Section: Fisher Vector and Tf-idfmentioning

confidence: 99%

Large-scale image retrieval with compressed Fisher vectors

Perronnin

Liu

Sánchez

et al. 2010

2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition

713

572

View full text Add to dashboard Cite

The problem of large-scale image search has been traditionally addressed with the bag-of-visual-words (BOV). In this article, we propose to use as an alternative the Fisher kernel framework. We first show why the Fisher representation is well-suited to the retrieval problem: it describes an image by what makes it different from other images. One drawback of the Fisher vector is that it is high-dimensional and, as opposed to the BOV, it is dense. The resulting memory and computational costs do not make Fisher vectors directly amenable to large-scale retrieval. Therefore, we compress Fisher vectors to reduce their memory footprint and speed-up the retrieval. We compare three binarization approaches: a simple approach devised for this representation and two standard compression techniques. We show on two publicly available datasets that compressed Fisher vectors perform very well using as little as a few hundreds of bits per image, and significantly better than a very recent compressed BOV approach.

show abstract

Section: Fisher Vector and Tf-idfmentioning

confidence: 99%

Large-scale image retrieval with compressed Fisher vectors

Perronnin

Liu

Sánchez

et al. 2010

2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition

713

572

View full text Add to dashboard Cite

show abstract

“…The Dirichlet Fisher kernel is also related to the Pólya Fisher kernel which has been first proposed in [7] for text categorization and then presented in [4] for visual classification. The Pólya model accounts for symbol (word) burstiness, which is measured discretely, by means of the compound of Dirichlet and multinomial models [30].…”

Section: Discussionmentioning

confidence: 99%

“…In contrast, the proposed method deals with various types of features into which those countable/uncountable histograms are L 1 -normalized. In addition, the Pólya method [7,4] inevitably requires to learn hyper parameters, which is computationally exhaustive in the case of large-scale high-dimensional histogram features.…”

Section: Discussionmentioning

confidence: 99%

Dirichlet-Based Histogram Feature Transform for Image Classification

Kobayashi

2014

2014 IEEE Conference on Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

Histogram-based features have significantly contributed to recent development of image classifications, such as by SIFT local descriptors. In this paper, we propose a method to efficiently transform those histogram features for improving the classification performance. The (L 1 -normalized) histogram feature is regarded as a probability mass function, which is modeled by Dirichlet distribution. Based on the probabilistic modeling, we induce the Dirichlet Fisher kernel for transforming the histogram feature vector. The method works on the individual histogram feature to enhance the discriminative power at a low computational cost. On the other hand, in the bag-of-feature (BoF) framework, the Dirichlet mixture model can be extended to Gaussian mixture by transforming histogram-based local descriptors, e.g., SIFT, and thereby we propose the method of Dirichlet-derived GMM Fisher kernel. In the experiments on diverse image classification tasks including recognition of subordinate objects and material textures, the proposed methods improve the performance of the histogrambased features and BoF-based Fisher kernel, being favorably competitive with the state-of-the-arts.

show abstract

“…The role of the Fisher information matrix, however, is less significant, as shown in [67], and then, can be approximated by the identity matrix. This approach has been used in [68], for instance, where a Fisher kernel was developed in the case of probabilistic latent semantic analysis, and in [69], where it was used to justify the success of the term frequency and inverse document frequency (TF-IDF) representation for documents. In the following, we shall derive the Fisher kernel for our model.…”

Section: Application To Svmsmentioning

confidence: 99%

A Model-Based Approach for Discrete Data Clustering and Feature Weighting Using MAP and Stochastic Complexity

Bouguila

2009

IEEE Trans. Knowl. Data Eng.

View full text Add to dashboard Cite

In this paper, we consider the problem of unsupervised discrete feature selection/weighting. Indeed, discrete data are an important component in many data mining, machine learning, image processing, and computer vision applications. However, much of the published work on unsupervised feature selection has concentrated on continuous data. We propose a probabilistic approach that assigns relevance weights to discrete features that are considered as random variables modeled by finite discrete mixtures. The choice of finite mixture models is justified by its flexibility which has led to its widespread application in different domains. For the learning of the model, we consider both Bayesian and information-theoretic approaches through stochastic complexity. Experimental results are presented to illustrate the feasibility and merits of our approach on a difficult problem which is clustering and recognizing visual concepts in different image data. The proposed approach is successfully applied also for text clustering.

show abstract

Deriving TF-IDF as a Fisher Kernel

Cited by 28 publications

References 5 publications

Large-scale image retrieval with compressed Fisher vectors

Large-scale image retrieval with compressed Fisher vectors

Dirichlet-Based Histogram Feature Transform for Image Classification

A Model-Based Approach for Discrete Data Clustering and Feature Weighting Using MAP and Stochastic Complexity

Contact Info

Product

Resources

About