Supervised Learning of Quantizer Codebooks by Information Loss Minimization

Authors:
Svetlana Lazebnik;Maxim Raginsky
Affiliations:
University of North Carolina at Chapel Hill, Chapel Hill;Duke University, Durham
Venue:
IEEE Transactions on Pattern Analysis and Machine Intelligence
Year:
2009

Citing 0
Cited 32

Descriptive visual words and visual phrases for image applications

MM '09 Proceedings of the 17th ACM international conference on Multimedia
Coherent phrase model for efficient image near-duplicate retrieval

IEEE Transactions on Multimedia
Hierarchical appearance-based classifiers for qualitative spatial localization

IROS'09 Proceedings of the 2009 IEEE/RSJ international conference on Intelligent robots and systems
Incremental indexing and distributed image search using shared randomized vocabularies

Proceedings of the international conference on Multimedia information retrieval
Multi-level pixel-based texture classification through efficient prototype selection via normalized cut

Pattern Recognition
The third eye: mining the visual cognition across multi-language communities

Proceedings of the international conference on Multimedia
Building contextual visual vocabulary for large-scale image applications

Proceedings of the international conference on Multimedia
Max-margin dictionary learning for multiclass image categorization

ECCV'10 Proceedings of the 11th European conference on Computer vision: Part IV
Building descriptive and discriminative visual codebook for large-scale image applications

Multimedia Tools and Applications
Image classification using spatial pyramid coding and visual word reweighting

ACCV'10 Proceedings of the 10th Asian conference on Computer vision - Volume Part III
Optimizing visual vocabularies using soft assignment entropies

ACCV'10 Proceedings of the 10th Asian conference on Computer vision - Volume Part IV
Supervised visual vocabulary with category information

ACIVS'11 Proceedings of the 13th international conference on Advanced concepts for intelligent vision systems
Letters: A unified supervised codebook learning framework for classification

Neurocomputing
Efficient and Effective Visual Codebook Generation Using Additive Kernels

The Journal of Machine Learning Research
Discriminative compact pyramids for object and scene recognition

Pattern Recognition
Learning semantic features for action recognition via diffusion maps

Computer Vision and Image Understanding
Semi-supervised classification using tree-based self-organizing maps

AI'11 Proceedings of the 24th international conference on Advances in Artificial Intelligence
Class consistent k-means: Application to face and action recognition

Computer Vision and Image Understanding
Modulating Shape Features by Color Attention for Object Recognition

International Journal of Computer Vision
Learning compact visual descriptor for low bit rate mobile landmark search

IJCAI'11 Proceedings of the Twenty-Second international joint conference on Artificial Intelligence - Volume Volume Three
Supervised learning of Gaussian mixture models for visual vocabulary generation

Pattern Recognition
Compact and adaptive spatial pyramids for scene recognition

Image and Vision Computing
On achieving semi-supervised pattern recognition by utilizing tree-based SOMs

Pattern Recognition
Hierarchical Classifiers for Robust Topological Robot Localization

Journal of Intelligent and Robotic Systems
Unsupervised and supervised visual codes with restricted boltzmann machines

ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part V
Transductive cost-sensitive lung cancer image classification

Applied Intelligence
Weakly supervised codebook learning by iterative label propagation with graph quantization

Signal Processing
Discriminative codebook learning for Web image search

Signal Processing
Approximate nearest neighbor search to support manual image annotation of large domain-specific datasets

Proceedings of the International Workshop on Video and Image Ground Truth in Computer Vision Applications
Joint learning and weighting of visual vocabulary for bag-of-feature based tissue classification

Pattern Recognition
Unsupervised approximate-semantic vocabulary learning for human action and video classification

Pattern Recognition Letters
Bilevel visual words coding for image classification

IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence

Quantified Score

Hi-index	0.15

Visualization

Abstract

This paper proposes a technique for jointly quantizing continuous features and the posterior distributions of their class labels based on minimizing empirical information loss such that the quantizer index of a given feature vector approximates a sufficient statistic for its class label. Informally, the quantized representation retains as much information as possible for classifying the feature vector correctly. We derive an alternating minimization procedure for simultaneously learning codebooks in the euclidean feature space and in the simplex of posterior class distributions. The resulting quantizer can be used to encode unlabeled points outside the training set and to predict their posterior class distributions, and has an elegant interpretation in terms of lossless source coding. The proposed method is validated on synthetic and real data sets and is applied to two diverse problems: learning discriminative visual vocabularies for bag-of-features image classification and image segmentation.