Color Equivariant Convolutional Networks

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

68 Downloads (Pure)

Abstract

Color is a crucial visual cue readily exploited by Convolutional Neural Networks (CNNs) for object recognition. However, CNNs struggle if there is data imbalance between color variations introduced by accidental recording conditions. Color invariance addresses this issue but does so at the cost of removing all color information, which sacrifices discriminative power. In this paper, we propose Color Equivariant Convolutions (CEConvs), a novel deep learning building block that enables shape feature sharing across the color spectrum while retaining important color information. We extend the notion of equivariance from geometric to photometric transformations by incorporating parameter sharing over hue-shifts in a neural network. We demonstrate the benefits of CEConvs in terms of downstream performance to various tasks and improved robustness to color changes, including train-test distribution shifts. Our approach can be seamlessly integrated into existing architectures, such as ResNets, and offers a promising solution for addressing color-based domain shifts in CNNs.
Original languageEnglish
Title of host publication37th Conference on Neural Information Processing Systems
Number of pages20
Publication statusPublished - 2023
Event37th Annual Conference on Neural Information Processing Systems - New Orleans, United States
Duration: 10 Dec 202316 Dec 2023
Conference number: 37

Conference

Conference37th Annual Conference on Neural Information Processing Systems
Abbreviated titleNeurIPS 2023
Country/TerritoryUnited States
CityNew Orleans
Period10/12/2316/12/23

Keywords

  • color equivariance
  • Equivariance
  • color robustness
  • equivariant convolutions

Fingerprint

Dive into the research topics of 'Color Equivariant Convolutional Networks'. Together they form a unique fingerprint.

Cite this