Request PDF on ResearchGate | Hierarchical Gaussianization for Image Classification | In this paper, we propose a new image representation to capture both. In this paper, we propose a new image representation to capture both the appearance and spatial information for image classification. Hierarchical Gaussianization for Image Classification. Xi Zhou.. cal Gaussianization, each image is represented by a Gaus-. please see the pdf file.

Author: Tegrel Faurr
Country: Lithuania
Language: English (Spanish)
Genre: Software
Published (Last): 28 August 2011
Pages: 34
PDF File Size: 8.79 Mb
ePub File Size: 4.56 Mb
ISBN: 941-2-55091-987-8
Downloads: 91046
Price: Free* [*Free Regsitration Required]
Uploader: Mikamuro

Hierarchical Gaussianization for image classification.

Download PDF Cite this paper. In this paper, we propose a new image representation to capture both the appearance and spatial information for image classification applications. First, we model the feature vectors, from the whole corpus, from each image and at each individual patch, in a Bayesian hierarchical framework using mixtures of Gaussians.

After such a hierarchical Gaussianization, each image is represented by a Gaussian mixture model GMM for its appearance, and several Gaussian maps for its spatial layout. Then we extract the appearance information from the GMM parameters, and the spatial information from global and local statistics over Gaussian maps.



Finally, we employ a supervised dimension reduction technique called DAP discriminant attribute projection to remove noise directions and to further enhance the discriminating power of our representation.

We justify that the traditional histogram representation and the spatial pyramid gaussianziation are special cases of our hierarchical Gaussianization. We compare our new representation with other approaches in scene classification, object recognition and face recognition, and our performance ranks among the top in all three tasks. Facial recognition system Computer vision Mathematics Histogram Mixture model Gaussian process Dimensionality reduction Contextual image classification Feature vector Machine learning Artificial intelligence Spatial analysis Pattern recognition.

A k-means clustering algorithm. Hartigan 1 Estimated H-index: Improving “bag – of – keypoints” image categorisation.

Farquhar 1 Estimated H-index: Cited 40 Source Add To Collection. Caltech Object Category Dataset. Gregory Griffin 2 Estimated H-index: Cited Source Add To Collection.

Adapted vocabularies for generic visual categorization. Florent Perronnin 43 Estimated H-index: Blei 58 Classificaiton H-index: Within-class covariance normalization for SVM-based speaker recognition. Hatch 4 Estimated Hierarcnical A K-Means Clustering Algorithm. Large scale discriminative training of hidden Markov models for speech recognition.

Woodland 48 Estimated H-index: Bernt Schiele 77 Estimated H-index: A GMM parts based face representation for improved verification through relevance adaptation. Simon Lucey 31 Estimated H-index: A practical view of large-scale classification: Ref Source Add To Collection. Unsupervised and supervised visual codes with restricted boltzmann machines.


Hierarchical Gaussianization for image classification

Hanlin Goh 7 Estimated H-index: Semantic image representation for visual recognition. Nuno Vasconcelos 51 Estimated H-index: Shrinkage Expansion Adaptive Metric Learning.

Qilong Wang 8 Estimated H-index: Learning hybrid part filters for scene recognition. Yingbin Zheng 7 Estimated H-index: Learning representative and discriminative image representation by deep appearance and spatial coding.

Bingyuan Liu 4 Estimated H-index: Gang Hua Stevens Institute of Technology.

Disruption-tolerant networking protocols and services for disaster response communication. Sarwar UddinYusuf. Efficient highly over-complete sparse coding using a mixture model.

Jianchao Yang 32 Estimated H-index: Spatially local coding for object recognition.

Sancho McCann 4 Estimated H-index: Other Papers By First Author. Real-world acoustic event detection pattern recognition letters [IF: Kuhl Rochester Institute of Technology. Lowe University of British Columbia. Beyond Bags of Features: Are you looking for