.png){kind=icon}
返回
Unsupervised Blur Kernel Estimation and Correction for Blind Super-Resolution
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2022.3170053.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Blind super-resolution (blind-SR) is an important task in the field of computer vision and has various applications in real-world. Blur kernel estimation is the main element of blind-SR along with the adaptive SR networks and a more accurately estimated kernel guarantees a better performance. Recently, generative adversarial networks (GANs), comparing recurrence patches across scales, have been the most successful unsupervised kernel estimation methods. However, they still involve several problems. (1) Their sharpness discrimination ability has been noted as being too weak, causing them to focus more on pattern shapes than sharpness. (2) In some cases, kernel correction processes were omitted; however, these are essential because the optimally generated kernel may be narrower than a point spread function (PSF) except when the PSF is ideal low-pass filter. (3) Previous studies also did not consider that GANs are affected by the thickness of edges as well as PSF. Thus, in this paper, 1) we propose a degradation and ranking comparison process designed to induce GAN models to became sensitive to image sharpness, and 2) propose a scale-free kernel correction technique using Gaussian kernel approximation including a thickness parameter. To improve the kernel accuracy further, we 3) propose a combination model of the proposed GAN and DIP(deep image prior) for more supervision, and designed a kernel correction network to propagate gradients through developed correction method. Several experiments demonstrate that our methods enhanced the l(2) error and the shape of the kernel significantly. In addition, by combining with ordinary blind-SR algorithms, the best reconstruction accuracy was achieved among unsupervised blur kernel estimation methods.
Keyword:
Blind super-resolution
kernel estimation
kernel correction
generative adversarial networks
deep image prior
期刊
IF:
3.6
论文数:
9.8W
被引数:
29.4W
机构
暂无机构信息
引用论文
In vivo DTI of the healthy and injured cat spinal cord at high spatial and angular resolution高空间和角度分辨率下健康和受伤的猫脊髓的体内DTI.png)
NeuroImage
NeuroImageIF0
PREAI
分享收藏Red, green, and blue electrochromism in ambipolar poly(amine–amide–imide)s based on electroactive tetraphenyl‐p‐phenylenediamine units基于电活性四苯基 p-苯二胺单元的双极性聚 (胺-酰胺-酰亚胺) 中的红色，绿色和蓝色电致变色
PREAI
分享收藏分享收藏Atomic oxygen distributions in the Venus thermosphere: Comparisons between Venus Express observations and global model simulationsIcarus
IcarusIF0
PREAI
分享收藏Who Knows, who Cares? Dementia Knowledge among Nurses, Care Workers, and Family members of People Living with Dementia谁知道，谁在乎？痴呆症患者的护士，护理人员和家庭成员的痴呆症知识
OAAI
OAAI分享收藏End-to-End Image Super-Resolution via Deep and Shallow Convolutional Networks通过深度和浅层卷积网络实现端到端图像超分辨率IEEE ACCESSOAAI
IEEE ACCESSIF3.6
OAAI分享收藏
查看更多内容