Building Deep Networks on Grassmann Manifolds

Abstract

Learning representations on Grassmann manifolds is popular in quite a few visual recognition tasks. In order to enable deep learning on Grassmann manifolds, this paper proposes a deep network architecture by generalizing the Euclidean network paradigm to Grassmann manifolds. In particular, we design full rank mapping layers to transform input Grassmannian data to more desirable ones, exploit re-orthonormalization layers to normalize the resulting matrices, study projection pooling layers to reduce the model complexity in the Grassmannian context, and devise projection mapping layers to respect Grassmannian geometry and meanwhile achieve Euclidean forms for regular output layers. To train the Grassmann networks, we exploit a stochastic gradient descent setting on manifolds of the connection weights, and study a matrix generalization of backpropagation to update the structured data. The evaluations on three visual recognition tasks show that our Grassmann networks have clear advantages over existing Grassmann learning methods, and achieve results comparable with state-of-the-art approaches.

AAAI'18 paper

Cited by in corpus (14)

• Recurrent Embedding Aggregation Network for Video Face Recognition
• Exchangeable deep neural networks for set-to-set matching and learning
• Coordinate Independent Convolutional Networks -- Isometry and Gauge Equivariant Convolutions on Riemannian Manifolds
• Revisiting 2D Convolutional Neural Networks for Graph-based Applications
• HAN: An Efficient Hierarchical Self-Attention Network for Skeleton-Based Gesture Recognition
• MVC-Net: A Convolutional Neural Network Architecture for Manifold-Valued Images With Applications
• Deep and Shallow Covariance Feature Quantization for 3D Facial Expression Recognition
• Constructing the Matrix Multilayer Perceptron and its Application to the VAE
• Skeleton-Based Hand Gesture Recognition by Learning SPD Matrices with Neural Networks
• Flow-based Generative Models for Learning Manifold to Manifold Mappings
• The flag manifold as a tool for analyzing and comparing data sets
• GeomNet: A Neural Network Based on Riemannian Geometries of SPD Matrix Space and Cholesky Space for 3D Skeleton-Based Interaction Recognition
• Graph-Preserving Grid Layout: A Simple Graph Drawing Method for Graph Classification using CNNs
• Grassmannian Discriminant Maps (GDM) for Manifold Dimensionality Reduction with Application to Image Set Classification