• Corpus ID: 243832769

Frequency-Aware Physics-Inspired Degradation Model for Real-World Image Super-Resolution

@article{Dong2021FrequencyAwarePD,
  title={Frequency-Aware Physics-Inspired Degradation Model for Real-World Image Super-Resolution},
  author={Zhenxing Dong and Hong Cao and Wang Shen and Yu Gan and Yuye Ling and Guangtao Zhai and Yikai Su},
  journal={ArXiv},
  year={2021},
  volume={abs/2111.03301}
}
Current learning-based single image super-resolution (SISR) algorithms underperform on real data due to the deviation in the assumed degradation process from that in the real-world scenario. Conventional degradation processes consider applying blur, noise, and downsampling (typically bicubic downsampling) on high-resolution (HR) images to synthesize lowresolution (LR) counterparts. However, few works on degradation modelling have taken the physical aspects of the optical imaging system into… 

Figures and Tables from this paper

Few-Shot Real Image Super-resolution via Distortion-Relation Guided Transfer Learning

TLDR
This paper is the first to investigate the few-shot real image super-resolution and propose a Distortion-Relation guided Transfer Learning (termed as DRTL) framework and instantiate DRTL integrated with pre-training and meta-learning pipelines as an embodiment to realize a distortion-relation aware FS-RSR.

References

SHOWING 1-10 OF 29 REFERENCES

Real-World Super-Resolution via Kernel Estimation and Noise Injection

TLDR
This work designs a novel degradation framework for real-world images by estimating various blur kernels as well as real noise distributions and proposes a real- world super-resolution model aiming at better perception.

Camera Lens Super-Resolution

TLDR
This paper investigates SR from the perspective of camera lenses, named as CameraSR, which aims to alleviate the intrinsic tradeoff between resolution (R) and field-of-view (V) in realistic imaging systems and quantitatively analyzes the performance of commonly-used synthetic degradation models.

Designing a Practical Degradation Model for Deep Blind Image Super-Resolution

TLDR
The experimental results demonstrate that the new degradation model can help to significantly improve the practicability of deep super-resolvers, thus providing a powerful alternative solution for real SISR applications.

Blind image super-resolution with spatially variant degradations

TLDR
This work proposes a solution that relies on a degradation aware SR network to synthesize the HR image given a low resolution image and the corresponding blur kernel and presents an optimization procedure that is able to recover both the degradation kernel and the high resolution image by minimizing the error predicted by the kernel discriminator.

Kernel Modeling Super-Resolution on Real Low-Resolution Images

TLDR
The proposed KMSR consists of two stages: a pool of realistic blur-kernels with a generative adversarial network (GAN) and then a super-resolution network with HR and corresponding LR images constructed with the generated kernels that incorporates blur-kernel modeling in the training.

Blind Super-Resolution With Iterative Kernel Correction

TLDR
An iterative correction scheme -- IKC that achieves better results than direct kernel estimation in blind SR problem and an effective SR network architecture using spatial feature transform (SFT) layers to handle multiple blur kernels, named SFTMD.

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

TLDR
SRGAN, a generative adversarial network (GAN) for image super-resolution (SR), is presented, to its knowledge, the first framework capable of inferring photo-realistic natural images for 4x upscaling factors and a perceptual loss function which consists of an adversarial loss and a content loss.

Perceptual Losses for Real-Time Style Transfer and Super-Resolution

TLDR
This work considers image transformation problems, and proposes the use of perceptual loss functions for training feed-forward networks for image transformation tasks, and shows results on image style transfer, where aFeed-forward network is trained to solve the optimization problem proposed by Gatys et al. in real-time.

Image super-resolution as sparse representation of raw image patches

TLDR
It is shown that a small set of randomly chosen raw patches from training images of similar statistical nature to the input image generally serve as a good dictionary, in the sense that the computed representation is sparse and the recovered high-resolution image is competitive or even superior in quality to images produced by other SR methods.

Super-resolution from a single image

TLDR
This paper proposes a unified framework for combining the classical multi-image super-resolution and the example-based super- resolution, and shows how this combined approach can be applied to obtain super resolution from as little as a single image (with no database or prior examples).