Stacks of convolutional Restricted Boltzmann Machines for shift-invariant feature learning
- 1 June 2009
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2735-2742
- https://doi.org/10.1109/cvpr.2009.5206577
Abstract
In this paper we present a method for learning class-specific features for recognition. Recently a greedy layer-wise procedure was proposed to initialize weights of deep belief networks, by viewing each layer as a separate restricted Boltzmann machine (RBM). We develop the convolutional RBM (C-RBM), a variant of the RBM model in which weights are shared to respect the spatial structure of images. This framework learns a set of features that can generate the images of a specific object class. Our feature extraction model is a four layer hierarchy of alternating filtering and maximum subsampling. We learn feature parameters of the first and third layers viewing them as separate C-RBMs. The outputs of our feature extraction hierarchy are then fed as input to a discriminative classifier. It is experimentally demonstrated that the extracted features are effective for object detection, using them to obtain performance comparable to the state of the art on handwritten digit recognition and pedestrian detection.Keywords
This publication has 19 references indexed in Scilit:
- Classification using intersection kernel support vector machines is efficientPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2008
- Pedestrian Detection via Classification on Riemannian ManifoldsIeee Transactions On Pattern Analysis and Machine Intelligence, 2008
- Classification using discriminative restricted Boltzmann machinesPublished by Association for Computing Machinery (ACM) ,2008
- A trainable feature extractor for handwritten digit recognitionPattern Recognition, 2007
- What makes a good model of natural images?Published by Institute of Electrical and Electronics Engineers (IEEE) ,2007
- POP: Patchwork of Parts Models for Object RecognitionInternational Journal of Computer Vision, 2007
- Reducing the Dimensionality of Data with Neural NetworksScience, 2006
- Fields of Experts: A Framework for Learning Image PriorsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Continuous restricted Boltzmann machine with an implementable training algorithmIEE Proceedings - Vision, Image, and Signal Processing, 2003
- Training Products of Experts by Minimizing Contrastive DivergenceNeural Computation, 2002