Loading...
Please wait, while we are loading the content...
Similar Documents
Learning convolutional feature hierarchies for visual recognition (2010)
| Content Provider | CiteSeerX |
|---|---|
| Author | Lecun, Yann Boureau, Y.-Lan Gregor, Karol Mathieu, Michaël Kavukcuoglu, Koray Sermanet, Pierre |
| Abstract | We propose an unsupervised method for learning multi-stage hierarchies of sparse convolutional features. While sparse coding has become an increasingly popular method for learning visual features, it is most often trained at the patch level. Applying the resulting filters convolutionally results in highly redundant codes because overlapping patches are encoded in isolation. By training convolutionally over large image windows, our method reduces the redudancy between feature vectors at neighboring locations and improves the efficiency of the overall representation. In addition to a linear decoder that reconstructs the image from sparse features, our method trains an efficient feed-forward encoder that predicts quasisparse features from the input. While patch-based training rarely produces anything but oriented edge detectors, we show that convolutional training produces highly diverse filters, including center-surround filters, corner detectors, cross detectors, and oriented grating detectors. We show that using these filters in multistage convolutional network architecture improves performance on a number of visual recognition and detection tasks. 1 |
| File Format | |
| Publisher Date | 2010-01-01 |
| Publisher Institution | In NIPS’10 |
| Access Restriction | Open |
| Subject Keyword | Unsupervised Method Feature Vector Redundant Code Neighboring Location Large Image Window Multistage Convolutional Network Architecture Patch Level Convolutional Training Sparse Convolutional Feature Efficient Feed-forward Encoder Linear Decoder Diverse Filter Detection Task Convolutional Feature Hierarchy Visual Feature Sparse Feature Multi-stage Hierarchy Edge Detector Sparse Coding Corner Detector Overall Representation Cross Detector Patch-based Training Popular Method Visual Recognition Center-surround Filter |
| Content Type | Text |
