Escuela Superior Politécnica del Litoral, ESPOL, Facultad de Ingeniería en Electricidad y Computación, CIDIS, Campus Gustavo Galindo Km. 30.5 Vía Perimetral, P.O. Box 09-01-5863, Guayaquil 090112, Ecuador.
Computer Vision Center, Edifici O, Campus UAB, Bellaterra, 08193 Barcelona, Spain.
Sensors (Basel). 2022 Mar 14;22(6):2254. doi: 10.3390/s22062254.
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.
本文提出了一种迁移域策略，以解决低分辨率热传感器的局限性，并生成具有合理质量的更高分辨率图像。所提出的技术采用 CycleGAN 架构，并在生成器中使用 ResNet 作为编码器，同时使用注意力模块和新的损失函数。该网络在使用三个不同热传感器获取的多分辨率热图像数据集上进行训练。结果报告在 2021 年 CVPR-PBVS 第二次热图像超分辨率挑战赛上的性能基准测试结果优于最先进的方法。这项工作的代码可在线获得。
