A Novel Domain Transfer-Based Approach for Unsupervised Thermal Image Super-Resolution

This paper presents a transfer domain strategy to tackle the limitations of low-resolution thermal sensors and generate higher-resolution images of reasonable quality. The proposed technique employs a CycleGAN architecture and uses a ResNet as an encoder in the generator along with an attention modu...

Descripción completa

Detalles Bibliográficos
Autores: Rivadeneira, Rafael E.|||0000-0002-5327-2048, Sappa, Angel|||0000-0003-2468-0031, Vintimilla, Boris X.|||0000-0001-8904-0209, Hammoud, Riad|||0000-0002-9007-0357
Tipo de recurso: artículo
Fecha de publicación:2022
País:España
Institución:Universitat Autònoma de Barcelona
Repositorio:Dipòsit Digital de Documents de la UAB
Idioma:inglés
OAI Identifier:oai:ddd.uab.cat:258167
Acceso en línea:https://ddd.uab.cat/record/258167
https://dx.doi.org/urn:doi:10.3390/s22062254
Access Level:acceso abierto
Palabra clave:Thermal image super-resolution
Unsupervised super-resolution
Thermal images
Attention module
Semiregistered thermal images
Descripción
Sumario:This paper presents a transfer domain strategy to tackle the limitations of low-resolution thermal sensors and generate higher-resolution images of reasonable quality. The proposed technique employs a CycleGAN architecture and uses a ResNet as an encoder in the generator along with an attention module and a novel loss function. The network is trained on a multi-resolution thermal image dataset acquired with three different thermal sensors. Results report better performance benchmarking results on the 2nd CVPR-PBVS-2021 thermal image super-resolution challenge than state-of-the-art methods. The code of this work is available online.