Abstract | ||
---|---|---|
We propose Eigen-RBM, a scalable Restricted Boltzmann Machine (RBM) for visual recognition in which the number of free parameters to learn is independent of the image size. Eigen-RBM exploits the global structure of the image and does not impose any locality or translation-invariance assumption, and regularizes the network weights to be a linear combination of a set of predefined filters. We show that, compared to basic RBM, Eigen-RBM can achieve similar or better performance in both recognition and sample generation with significantly less training time. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1109/ICIP.2014.7025557 | Image Processing |
Keywords | Field | DocType |
Boltzmann machines,eigenvalues and eigenfunctions,image filtering,image recognition,learning (artificial intelligence),eigen-RBM,image global structure,image size,network weights,predefined filters,scalable learning,scalable restricted Boltzmann machine,visual recognition,Feature Extraction,Generative Models,Image Classification,Machine Learning,Restricted Boltzmann Machines | Restricted Boltzmann machine,Linear combination,Locality,Boltzmann machine,Pattern recognition,Computer science,Feature extraction,Artificial intelligence,Contextual image classification,Feature learning,Scalability | Conference |
ISSN | Citations | PageRank |
1522-4880 | 0 | 0.34 |
References | Authors | |
17 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Elnaz Barshan | 1 | 0 | 1.01 |
Paul W. Fieguth | 2 | 612 | 54.17 |