Learning a robust representation via a deep network on symmetric positive definite manifolds
作者：
Highlights：
• We formulate the convolutional feature aggregation as an SPD matrix non-linear generation and transformation problem on the Riemannian manifold.
• Our model can be implemented under an end-to-end learning framework.
• We design three novel layers to implement the convolutional feature aggregation.
• We exploit the faster matrix operation to speed up the computation, and the component decomposition and retraction of the orthogonal Stiefel manifold to carry out the backpropagation.
• Our approach notably outperforms the state-of-the-art methods.
摘要
•We formulate the convolutional feature aggregation as an SPD matrix non-linear generation and transformation problem on the Riemannian manifold.•Our model can be implemented under an end-to-end learning framework.•We design three novel layers to implement the convolutional feature aggregation.•We exploit the faster matrix operation to speed up the computation, and the component decomposition and retraction of the orthogonal Stiefel manifold to carry out the backpropagation.•Our approach notably outperforms the state-of-the-art methods.
论文关键词：Feature aggregation,SPD Matrix,Riemannian manifold,Deep convolutional network
论文评审过程：Received 5 April 2018, Revised 2 January 2019, Accepted 13 March 2019, Available online 14 March 2019, Version of Record 18 March 2019.
论文官网地址：https://doi.org/10.1016/j.patcog.2019.03.007
