Second-order Convolutional Neural Networks

03/20/2017
by   Kaicheng Yu, et al.
0

Convolutional Neural Networks (CNNs) have been successfully applied to many computer vision tasks, such as image classification. By performing linear combinations and element-wise nonlinear operations, these networks can be thought of as extracting solely first-order information from an input image. In the past, however, second-order statistics computed from handcrafted features, e.g., covariances, have proven highly effective in diverse recognition tasks. In this paper, we introduce a novel class of CNNs that exploit second-order statistics. To this end, we design a series of new layers that (i) extract a covariance matrix from convolutional activations, (ii) compute a parametric, second-order transformation of a matrix, and (iii) perform a parametric vectorization of a matrix. These operations can be assembled to form a Covariance Descriptor Unit (CDU), which replaces the fully-connected layers of standard CNNs. Our experiments demonstrate the benefits of our new architecture, which outperform the first-order CNNs, while relying on up to 90 fewer parameters.

READ FULL TEXT
research
11/29/2018

Global Second-order Pooling Neural Networks

Deep Convolutional Networks (ConvNets) are fundamental to, besides large...
research
11/29/2018

Global Second-order Pooling Convolutional Networks

Deep Convolutional Networks (ConvNets) are fundamental to, besides large...
research
01/23/2018

Statistically Motivated Second Order Pooling

Second-order pooling, a.k.a. bilinear pooling, has proven effective for ...
research
06/20/2016

Learning Convolutional Neural Networks using Hybrid Orthogonal Projection and Estimation

Convolutional neural networks (CNNs) have yielded the excellent performa...
research
03/20/2017

SORT: Second-Order Response Transform for Visual Recognition

In this paper, we reveal the importance and benefits of introducing seco...
research
04/30/2023

Second-order Anisotropic Gaussian Directional Derivative Filters for Blob Detection

Interest point detection methods have received increasing attention and ...
research
02/09/2016

Parameterizing Region Covariance: An Efficient Way To Apply Sparse Codes On Second Order Statistics

Sparse representations have been successfully applied to signal processi...

Please sign up or login with your details

Forgot password? Click here to reset