Abstract | ||
---|---|---|
Unsupervised feature learning has shown impressive results for a wide range of input modalities, in particular for object classification tasks in computer vision. Using a large amount of unlabeled data, unsupervised feature learning methods are utilized to construct high-level representations that are discriminative enough for subsequently trained supervised classification algorithms. However, it has never been \emph{quantitatively} investigated yet how well unsupervised learning methods can find \emph{low-level representations} for image patches without any additional supervision. In this paper we examine the performance of pure unsupervised methods on a low-level correspondence task, a problem that is central to many Computer Vision applications. We find that a special type of Restricted Boltzmann Machines (RBMs) performs comparably to hand-crafted descriptors. Additionally, a simple binarization scheme produces compact representations that perform better than several state-of-the-art descriptors. |
Year | Venue | Field |
---|---|---|
2013 | CoRR | Modalities,Boltzmann machine,Semi-supervised learning,Pattern recognition,Computer science,Unsupervised learning,Artificial intelligence,Visual descriptors,Statistical classification,Discriminative model,Machine learning,Feature learning |
DocType | Volume | Citations |
Journal | abs/1301.2840 | 2 |
PageRank | References | Authors |
0.39 | 25 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Christian Osendorfer | 1 | 125 | 13.24 |
Justin Bayer | 2 | 157 | 32.38 |
Sebastian Urban | 3 | 22 | 5.38 |
P. Patrick Van Der Smagt | 4 | 274 | 35.19 |