Skip to main navigation Skip to search Skip to main content

Multiple adverse weather removal using adversarial and contrastive learning

  • Harbin Institute of Technology Shenzhen
  • School of Computer Science and Technology, Harbin Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Adverse weather removal is significant and valuable for real-world computer vision systems. Most existing algorithms focus on removing one specific type of weather and rely on paired synthesized weather datasets for training. In this paper, to train a multiple weather removal model with unpaired datasets, we introduce multi-dimension contrastive learning mechanism into generative adversarial networks. Specifically, our method consists of two blocks: multi-channel cycle translation (MCCT) block and multi-dimension contrastive learning (MDCL) block. MCCT is used to preserve the color and structural information and gain informative supervisory signals. MDCL is proposed to discriminate feature distribution among different image domains. Results of extensive experiments show that our method is superior over existing unsupervised methods and maintains a decent performance degradation when adapted to multiple weather removal tasks.

Original languageEnglish
Title of host publicationProceedings - 2023 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing and Data Security, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PCDS/Metaverse 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350319804
DOIs
StatePublished - 2023
Externally publishedYes
Event9th IEEE Smart World Congress, SWC 2023 - Portsmouth, United Kingdom
Duration: 28 Aug 202331 Aug 2023

Publication series

NameProceedings - 2023 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing and Data Security, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PCDS/Metaverse 2023

Conference

Conference9th IEEE Smart World Congress, SWC 2023
Country/TerritoryUnited Kingdom
CityPortsmouth
Period28/08/2331/08/23

Keywords

  • adverse weather removal
  • contrastive learning
  • deep learning
  • generative adversarial networks
  • unsupervised learning

Fingerprint

Dive into the research topics of 'Multiple adverse weather removal using adversarial and contrastive learning'. Together they form a unique fingerprint.

Cite this