Information theoretic learning for pixel-based visual agents

Authors:
Marco Gori;Stefano Melacci;Marco Lippi;Marco Maggini
Affiliations:
Department of Information Engineering, University of Siena, Siena, Italy;Department of Information Engineering, University of Siena, Siena, Italy;Department of Information Engineering, University of Siena, Siena, Italy;Department of Information Engineering, University of Siena, Siena, Italy
Venue:
ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part VI
Year:
2012

Citing 6
Cited 0

Distinctive Image Features from Scale-Invariant Keypoints

International Journal of Computer Vision
Learning generative visual models from few training examples: An incremental Bayesian approach tested on 101 object categories

Computer Vision and Image Understanding
Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations

ICML '09 Proceedings of the 26th Annual International Conference on Machine Learning
Visual Word Ambiguity

IEEE Transactions on Pattern Analysis and Machine Intelligence
Information Theoretic Learning: Renyi's Entropy and Kernel Perspectives

Information Theoretic Learning: Renyi's Entropy and Kernel Perspectives
Kernel Methods for Minimum Entropy Encoding

ICMLA '11 Proceedings of the 2011 10th International Conference on Machine Learning and Applications and Workshops - Volume 01

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we promote the idea of using pixel-based models not only for low level vision, but also to extract high level symbolic representations. We use a deep architecture which has the distinctive property of relying on computational units that incorporate classic computer vision invariances and, especially, the scale invariance. The learning algorithm that is proposed, which is based on information theory principles, develops the parameters of the computational units and, at the same time, makes it possible to detect the optimal scale for each pixel. We give experimental evidence of the mechanism of feature extraction at the first level of the hierarchy, which is very much related to SIFT-like features. The comparison shows clearly that, whenever we can rely on the massive availability of training data, the proposed model leads to better performances with respect to SIFT.