English
Related papers

Related papers: Equivariance versus Augmentation for Spherical Ima…

200 papers

Spherical data is found in many applications. By modeling the discretized sphere as a graph, we can accommodate non-uniformly distributed, partial, and changing samplings. Moreover, graph convolutions are computationally more efficient than…

Machine Learning · Computer Science 2019-04-11 Michaël Defferrard , Nathanaël Perraudin , Tomasz Kacprzak , Raphael Sgier

Convolutional neural networks (CNNs) have been widely used in various vision tasks, e.g. image classification, semantic segmentation, etc. Unfortunately, standard 2D CNNs are not well suited for spherical signals such as panorama images or…

Computer Vision and Pattern Recognition · Computer Science 2022-09-05 Yuqi Liu , Yin Wang , Haikuan Du , Shen Cai

Convolutional Neural Networks (CNN) have been pivotal to the success of many state-of-the-art classification problems, in a wide variety of domains (for e.g. vision, speech, graphs and medical imaging). A commonality within those domains is…

Machine Learning · Computer Science 2019-12-02 Rohan Ghosh , Anupam K. Gupta , Mehul Motani

Despite the importance of image representations such as histograms of oriented gradients and deep Convolutional Neural Networks (CNN), our theoretical understanding of them remains limited. Aiming at filling this gap, we investigate three…

Computer Vision and Pattern Recognition · Computer Science 2015-06-23 Karel Lenc , Andrea Vedaldi

The effectiveness of Convolutional Neural Networks (CNNs) has been substantially attributed to their built-in property of translation equivariance. However, CNNs do not have embedded mechanisms to handle other types of transformations. In…

Computer Vision and Pattern Recognition · Computer Science 2020-02-07 Ivan Sosnovik , Michał Szmaja , Arnold Smeulders

Learning equivariant representations is a promising way to reduce sample and model complexity and improve the generalization performance of deep neural networks. The spherical CNNs are successful examples, producing SO(3)-equivariant…

Computer Vision and Pattern Recognition · Computer Science 2020-10-27 Carlos Esteves , Ameesh Makadia , Kostas Daniilidis

Group Convolutional Neural Networks (G-CNNs) constrain learned features to respect the symmetries in the selected group, and lead to better generalization when these symmetries appear in the data. If this is not the case, however,…

Computer Vision and Pattern Recognition · Computer Science 2023-01-18 David W. Romero , Suhas Lohit

Convolutional neural networks (CNNs) are essential tools for computer vision tasks, but they lack traditionally desired properties of extracted features that could further improve model performance, e.g., rotational equivariance. Such…

Computer Vision and Pattern Recognition · Computer Science 2025-01-17 Yuexi Du , Jiazhen Zhang , Tal Zeevi , Nicha C. Dvornek , John A. Onofrey

An important goal in visual recognition is to devise image representations that are invariant to particular transformations. In this paper, we address this goal with a new type of convolutional neural network (CNN) whose invariance is…

Computer Vision and Pattern Recognition · Computer Science 2015-01-08 Julien Mairal , Piotr Koniusz , Zaid Harchaoui , Cordelia Schmid

In this work we investigate how to achieve equivariance to input transformations in deep networks, purely from data, without being given a model of those transformations. Convolutional Neural Networks (CNNs), for example, are equivariant to…

Computer Vision and Pattern Recognition · Computer Science 2021-11-29 Jianbo Jiao , João F. Henriques

Omnidirectional images and spherical representations of $3D$ shapes cannot be processed with conventional 2D convolutional neural networks (CNNs) as the unwrapping leads to large distortion. Using fast implementations of spherical and…

Computer Vision and Pattern Recognition · Computer Science 2020-12-09 Suhas Lohit , Shubhendu Trivedi

Spherical signals exist in many applications, e.g., planetary data, LiDAR scans and digitalization of 3D objects, calling for models that can process spherical data effectively. It does not perform well when simply projecting spherical data…

Computer Vision and Pattern Recognition · Computer Science 2022-01-17 Zhengyang Shen , Tiancheng Shen , Zhouchen Lin , Jinwen Ma

Many classes of images exhibit rotational symmetry. Convolutional neural networks are sometimes trained using data augmentation to exploit this, but they are still required to learn the rotation equivariance properties from the data.…

Machine Learning · Computer Science 2016-05-27 Sander Dieleman , Jeffrey De Fauw , Koray Kavukcuoglu

Little is known about the training dynamics of equivariant neural networks, in particular how it compares to data augmented training of their non-equivariant counterparts. Recently, neural tangent kernels (NTKs) have emerged as a powerful…

Machine Learning · Computer Science 2025-02-03 Philipp Misof , Pan Kessel , Jan E. Gerken

The success of convolutional networks in learning problems involving planar signals such as images is due to their ability to exploit the translation symmetry of the data distribution through weight sharing. Many areas of science and…

Machine Learning · Computer Science 2019-04-23 Taco Cohen , Mario Geiger , Jonas Köhler , Max Welling

This paper demonstrates spherical convolutional neural networks (S-CNN) offer distinct advantages over conventional fully-connected networks (FCN) at estimating scalar parameters of tissue microstructure from diffusion MRI (dMRI). Such…

Image and Video Processing · Electrical Eng. & Systems 2022-08-17 Tobias Goodwin-Allcock , Jason McEwen , Robert Gray , Parashkev Nachev , Hui Zhang

Convolutional Neural Networks (CNNs) traditionally encode translation equivariance via the convolution operation. Generalization to other transformations has recently received attraction to encode the knowledge of the data geometry in group…

Computer Vision and Pattern Recognition · Computer Science 2018-10-17 Vincent Andrearczyk , Adrien Depeursinge

Convolutional neural networks (CNNs) have achieved state-of-the-art results on many visual recognition tasks. However, current CNN models still exhibit a poor ability to be invariant to spatial transformations of images. Intuitively, with…

Computer Vision and Pattern Recognition · Computer Science 2019-12-04 Xu Shen , Xinmei Tian , Anfeng He , Shaoyan Sun , Dacheng Tao

The ability of convolutional neural networks (CNNs) to recognize objects regardless of their position in the image is due to the translation-equivariance of the convolutional operation. Group-equivariant CNNs transfer this equivariance to…

Computer Vision and Pattern Recognition · Computer Science 2023-04-13 Thomas Wimmer , Vladimir Golkov , Hoai Nam Dang , Moritz Zaiss , Andreas Maier , Daniel Cremers

Rotation-invariance is a desired property of machine-learning models for medical image analysis and in particular for computational pathology applications. We propose a framework to encode the geometric structure of the special Euclidean…

Computer Vision and Pattern Recognition · Computer Science 2020-02-21 Maxime W. Lafarge , Erik J. Bekkers , Josien P. W. Pluim , Remco Duits , Mitko Veta