Abstract:Sparse representation is a mathematical model for data representation that has proved to be a powerful tool for solving problems in various fields such as pattern recognition, machine learning, and computer vision. As one of the building blocks of the sparse representation method, dictionary learning plays an important role in the minimization of the reconstruction error between the original signal and its sparse representation in the space of the learned dictionary. Although using training samples directly as dictionary bases can achieve good performance, the main drawback of this method is that it may result in a very large and inefficient dictionary due to noisy training instances. To obtain a smaller and more representative dictionary, in this paper, we propose an approach called Laplacian sparse dictionary (LSD) learning. Our method is based on manifold learning and double sparsity. We incorporate the Laplacian weighted graph in the sparse representation model and impose the l1-norm sparsity on the dictionary. An LSD is a sparse overcomplete dictionary that can preserve the intrinsic structure of the data and learn a smaller dictionary for each class. The learned LSD can be easily integrated into a classification framework based on sparse representation. We compare the proposed method with other methods using three benchmark-controlled face image databases, Extended Yale B, ORL, and AR, and one uncontrolled person image dataset, i-LIDS-MA. Results show the advantages of the proposed LSD algorithm over state-of-the-art sparse representation based classification methods.

Locality-Constrained Sparse Auto-Encoder for Image Classification

Convolutional Sparse Autoencoders for Image Classification.

Representation Learning Via a Semi-Supervised Stacked Distance Autoencoder for Image Classification

Image classification using label constrained sparse coding

Laplacian Sparse Dictionary Learning for Image Classification Based on Sparse Representation

Low-Rank Sparse Coding for Image Classification

Non-Negativity and Locality Constrained Laplacian Sparse Coding for Image Classification

HSAE: A Hessian Regularized Sparse Auto-Encoders.

Fast Locality-Constrained Low-Rank Coding for Image Classification

Image Classification by Non-Negative Sparse Coding, Low-Rank and Sparse Decomposition

Laplacian Regularized Locality-Constrained Coding for Image Classification.

Locality Sensitive Dictionary Learning for Image Classification

Local Manifold Sparse Model for Image Classification.

Fine-Grained Image Classification Via Low-Rank Sparse Coding with General and Class-Specific Codebooks.

Sparse Autoencoder for Social Image Understanding.

Locality Constrained Encoding of Frequency and Spatial Information for Image Classification

Boosting Sparsity-Induced Autoencoder: A Novel Sparse Feature Ensemble Learning for Image Classification

LMAE: A Large Margin Auto-Encoders for Classification

Optimized Laplacian Sparse Coding for Image Classification.

Sparse Representation Learning of Data by Autoencoders with L_1/2 Regularization

Efficient Feature Coding Based On Auto-Encoder Network For Image Classification