%0 Journal Article
%T A Novel Domain Transfer-Based Approach for Unsupervised Thermal Image Super-Resolution
%A Rafael E. Rivadeneira
%A Angel Sappa
%A Boris X. Vintimilla
%A Riad I. Hammoud
%J Sensors
%D 2022
%V 22
%N 6
%F Rafael E. Rivadeneira2022
%O MSIAU;
%O exported from refbase (http://refbase.cvc.uab.es/show.php?record=3688), last updated on Tue, 25 Apr 2023 15:33:39 +0200
%X 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.
%K Thermal image super-resolution
%K unsupervised super-resolution
%K thermal images
%K attention module
%K semiregistered thermal images
%U http://refbase.cvc.uab.es/files/RSV2022b.pdf
%U http://dx.doi.org/10.3390/s22062254
%P 2254