Abstract:Image representation has been intensively explored in the domain of computer vision for its significant influence on the relative tasks such as image clustering and classification. It is valuable to learn a low-dimensional representation of an image which preserves its inherent information from the original image space. At the perspective of manifold learning, this is implemented with the local invariant idea to capture the intrinsic low-dimensional manifold embedded in the high-dimensional input space. Inspired by the recent successes of deep architectures, we propose a local invariant deep nonlinear mapping algorithm, called graph regularized auto-encoder (GAE). With the graph regularization, the proposed method preserves the local connectivity from the original image space to the representation space, while the stacked auto-encoders provide explicit encoding model for fast inference and powerful expressive capacity for complex modeling. Theoretical analysis shows that the graph regularizer penalizes the weighted Frobenius norm of the Jacobian matrix of the encoder mapping, where the weight matrix captures the local property in the input space. Furthermore, the underlying effects on the hidden representation space are revealed, providing insightful explanation to the advantage of the proposed method. Finally, the experimental results on both clustering and classification tasks demonstrate the effectiveness of our GAE as well as the correctness of the proposed theoretical analysis, and it also suggests that GAE is a superior solution to the current deep representation learning techniques comparing with variant auto-encoders and existing local invariant methods.

Regularized direct Linear Graph Embedding

Locality Pursuit Embedding

Image retrieval using dimensionality reduction

Graph Learning in Low Dimensional Space for Graph Convolutional Networks

Graph Embedding with Constraints

Low-rank 2D Local Discriminant Graph Embedding for Robust Image Feature Extraction

Graph Regularized Encoder-Decoder Networks for Image Representation Learning

A Generalized Least-Squares Approach Regularized with Graph Embedding for Dimensionality Reduction

Learnable Graph-Regularization for Matrix Decomposition

Signed Laplacian Embedding for Supervised Dimension Reduction.

Maximal Linear Embedding for Dimensionality Reduction

Global structure-guided neighborhood preserving embedding for dimensionality reduction

By Normalizing To Improve Generalized Foley-Sammon Transform In High-Dimensional Spaces-----With Application To Face Recognition

An Iterative Locally Linear Embedding Algorithm

A New Low-Rank Structurally Incoherent Algorithm for Robust Image Feature Extraction

Graph Regularized Auto-Encoders for Image Representation

Linear Manifold Regularization with Adaptive Graph for Semi-Supervised Dimensionality Reduction

Weighted Linear Embedding: Utilizing Local and Nonlocal Information Sufficiently

A framework of joint graph embedding and sparse regression for dimensionality reduction.

Face Recognition Base On Uncorrelated Linear Extension Of Graph Embedding

Fast and Flexible Large Graph Embedding Based on Anchors