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

Rivadeneira, Rafael E.; Sappa, Ángel D.; Vintimilla, Boris X.; Hammoud, Riad I.

doi:10.3390/s22062254

Cited by 6 publications

(8 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We compare our method with several competitive unsupervised image SR methods, including Bulat et al [ 28 ], FSSR [ 15 ], DASR [ 17 ], and Rivadeneira et al [ 26 ]. Among them, Bulat et al and Rivadeneira et al are for collaborative training, and FSSR and DASR are for two-step training.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…However, this method faces challenges in the restoration of images with severely damaged details. CinCGAN [ 25 ] and unsupervisedThSR [ 26 ] use the framework of CycleGAN [ 27 ] to learn the mapping of LR images to HR images using unpaired datasets. In these networks, cycle-consistency constraint is over-emphasized, resulting in inadequate training stability.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Frequency-Aware Degradation Modeling for Real-World Thermal Image Super-Resolution

Qu,

Chen,

et al. 2024

Entropy

View full text Add to dashboard Cite

The supervised super-resolution (SR) methods based on simple degradation assumptions (e.g., bicubic downsampling) have unsatisfactory generalization ability on real-world thermal images. To enhance the SR effect of real-world sceneries, we introduce an unsupervised SR framework for thermal images, incorporating degradation modeling and corresponding SR. Inspired by the physical prior that high frequency affects details and low frequency affects thermal contrast, we propose a frequency-aware degradation model, named TFADGAN. The model achieves image quality migration between thermal detectors of different resolutions by degrading different frequency components of the image from high-resolution (HR) to low-resolution (LR). Specifically, by adversarial learning with unpaired LR thermal images, the complex degradation processes of HR thermal images at low and high frequencies are modeled separately. Benefiting from the thermal characteristics mined from real-world images, the degraded images generated by TFADGAN are similar to LR thermal ones in terms of detail and contrast. Then, the SR model is trained based on the pseudo-paired data consisting of degraded images and HR images. Extensive experimental results demonstrate that the degraded images generated by TFADGAN provide reliable alternatives to real-world LR thermal images. In real-world thermal image experiments, the proposed SR framework can improve the peak signal-to-noise ratio (PSNR) and structural similarity degree (SSIM) by 1.28 dB and 0.02, respectively.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Frequency-Aware Degradation Modeling for Real-World Thermal Image Super-Resolution

Qu,

Chen,

et al. 2024

Entropy

View full text Add to dashboard Cite

show abstract

“…Deep learning methods include CNN-based methods [11][12][13][14][15][16][17][18] and GAN-based methods. [19][20][21][22][23][24][25] CNN-based methods are used for two major fields. Firstly, CNN can learn nonlinear mappings without indication because of its powerful fitting ability.…”

Section: Deep Learning Methodsmentioning

confidence: 99%

“…Then researchers tried to design GAN models specifically for infrared images by module improvement and introducing extra information. Rivadeneira et al 21 used the CycleGAN structure and employed ResNet as the generator's encoder. Qing.L., et al 22 used a module based on the channel attention mechanism in the generator at their work.…”

Section: Deep Learning Methodsmentioning

confidence: 99%

Infrared image super-resolution enhancement technology based on improved real-ESRGAN

Chen,

Yang,

Hao

et al. 2023

AOPC 2023: AI in Optics and Photonics

View full text Add to dashboard Cite

“…There are many image super-resolution methods such as [ 8 , 9 , 10 , 11 , 12 , 13 ]. For instance, ref.…”

Section: Related Workmentioning

confidence: 99%

Lightweight Super-Resolution with Self-Calibrated Convolution for Panoramic Videos

Shang

Liu

et al. 2022

Sensors

View full text Add to dashboard Cite

Panoramic videos are shot by an omnidirectional camera or a collection of cameras, and can display a view in every direction. They can provide viewers with an immersive feeling. The study of super-resolution of panoramic videos has attracted much attention, and many methods have been proposed, especially deep learning-based methods. However, due to complex architectures of all the methods, they always result in a large number of hyperparameters. To address this issue, we propose the first lightweight super-resolution method with self-calibrated convolution for panoramic videos. A new deformable convolution module is designed first, with self-calibration convolution, which can learn more accurate offset and enhance feature alignment. Moreover, we present a new residual dense block for feature reconstruction, which can significantly reduce the parameters while maintaining performance. The performance of the proposed method is compared to those of the state-of-the-art methods, and is verified on the MiG panoramic video dataset.

show abstract

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

Cited by 6 publications

References 53 publications

Frequency-Aware Degradation Modeling for Real-World Thermal Image Super-Resolution

Frequency-Aware Degradation Modeling for Real-World Thermal Image Super-Resolution

Infrared image super-resolution enhancement technology based on improved real-ESRGAN

Lightweight Super-Resolution with Self-Calibrated Convolution for Panoramic Videos

Contact Info

Product

Resources

About