Blind Image Super-Resolution: A Survey and Beyond

Liu, Anran; Liu, Yihao; Gu, Jinjin; Qiao, Yu; Chen, Dong

doi:10.48550/arxiv.2107.03055

Cited by 18 publications

(26 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Larger degradation space grants these models better generalization abilities. But compared with the huge degradation space in real scenarios, the variety of predefined degradations is still limited and these methods still fail in most applications [25,42,5].…”

Section: Predefined-degradation-basedmentioning

confidence: 99%

Learning the Degradation Distribution for Blind Image Super-Resolution

Luo¹,

Huang²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

Synthetic high-resolution (HR) & low-resolution (LR) pairs are widely used in existing super-resolution (SR) methods. To avoid the domain gap between synthetic and test images, most previous methods try to adaptively learn the synthesizing (degrading) process via a deterministic model. However, some degradations in real scenarios are stochastic and cannot be determined by the content of the image. These deterministic models may fail to model the random factors and content-independent parts of degradations, which will limit the performance of the following SR models. In this paper, we propose a probabilistic degradation model (PDM), which studies the degradation D as a random variable, and learns its distribution by modeling the mapping from a priori random variable z to D. Compared with previous deterministic degradation models, PDM could model more diverse degradations and generate HR-LR pairs that may better cover the various degradations of test images, and thus prevent the SR model from over-fitting to specific ones. Extensive experiments have demonstrated that our degradation model can help the SR model achieve better performance on different datasets.

show abstract

Section: Predefined-degradation-basedmentioning

confidence: 99%

Learning the Degradation Distribution for Blind Image Super-Resolution

Luo¹,

Huang²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Since the seminal work of SRCNN [6], many convolutional neural network (CNN) based SISR methods [7][8][9][10][11] have been proposed, most of which assume a pre-defined degradation process (e.g., bicubic downsampling) from HR to LR images. Despite the great success, the performance of these non-blind SISR methods will be much deteriorated when facing real-world images [12] because of the mismatch of degradation models between the training data and the real-world test data [13].…”

Section: Introductionmentioning

confidence: 99%

Efficient and Degradation-Adaptive Network for Real-World Image Super-Resolution

Liu¹,

Zhong²,

Zhang³

2022

Preprint

View full text Add to dashboard Cite

Efficient and effective real-world image super-resolution (Real-ISR) is a challenging task due to the unknown complex degradation of real-world images and the limited computation resources in practical applications. Recent research on Real-ISR has achieved significant progress by modeling the image degradation space; however, these methods largely rely on heavy backbone networks and they are inflexible to handle images of different degradation levels. In this paper, we propose an efficient and effective degradation-adaptive super-resolution (DASR) network, whose parameters are adaptively specified by estimating the degradation of each input image. Specifically, a tiny regression network is employed to predict the degradation parameters of the input image, while several convolutional experts with the same topology are jointly optimized to specify the network parameters via a non-linear mixture of experts. The joint optimization of multiple experts and the degradation-adaptive pipeline significantly extend the model capacity to handle degradations of various levels, while the inference remains efficient since only one adaptively specified network is used for super-resolving the input image. Our extensive experiments demonstrate that the proposed DASR is not only much more effective than existing methods on handling real-world images with different degradation levels but also efficient for easy deployment. Codes, models and datasets are available at https://github.com/csjliang/DASR.

show abstract

“…Yet, they are limited to the degradations within training datasets, and could not generalize well to out-of-distribution images. Readers are encouraged to refer to a recent blind SR survey [26] for a more comprehensive taxonomy.…”

Section: Introductionmentioning

confidence: 99%