当前位置: X-MOL 学术J. Electron. Imaging › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Perceptual-based super-resolution reconstruction using image-specific degradation estimation
Journal of Electronic Imaging ( IF 1.0 ) Pub Date : 2021-05-01 , DOI: 10.1117/1.jei.30.3.033007
Kawther Aarizou 1 , Abdelhamid Loukil 1
Affiliation  

Supervised single-image super-resolution (SISR) reconstruction models are trained with both low-resolution images (ILR) and their corresponding high-resolution images (IHR). During the training process, ILR are obtained by performing a bicubic downscaling on their IHR counterparts. This means that the model learns an inverted version of the bicubic downscaling, resulting in less realistic images that are limited to specific conditions. Generating realistic textures is non-trivial. The obtained details are either blurred or not reminiscent of the usually observed textures. SISR reconstruction with faithful ground-truth texture and no external information remains an issue, especially when the degradation model is not defined. We propose an unsupervised internal learning method of a small convolutional neural network (CNN) using the internal image statistics. We use the power of deep generative models to capture latent representation of patches within the test image across two scales and train a downscaling CNN Dw to learn how to downscale the image by matching these latent distributions. Dw constitutes the downscaling operation with the correct image-specific degradation and is subsequently used in the generation of the training dataset. Obtained results show the effectiveness of our image-degradation estimation method in extracting inner-image statistics for a better super-resolution perceptual reconstruction.

中文翻译:

基于图像的退化估计的基于感知的超分辨率重建

使用低分辨率图像(ILR)及其对应的高分辨率图像(IHR)训练有监督的单图像超分辨率(SISR)重建模型。在培训过程中,可以通过对IHR对应对象执行双三次缩减来获得ILR。这意味着该模型学习了三次三次降尺度的倒置形式,从而产生了受限于特定条件的不太真实的图像。生成逼真的纹理并非易事。所获得的细节模糊不清或无法让人联想到通常观察到的纹理。具有忠实的地面真实纹理并且没有外部信息的SISR重建仍然是一个问题,尤其是在未定义退化模型的情况下。我们提出了一种使用内部图像统计数据的小型卷积神经网络(CNN)的无监督内部学习方法。我们使用深度生成模型的功能来捕获测试图像中跨两个尺度的补丁的潜在表示,并训练降尺度的CNN Dw,以了解如何通过匹配这些潜在分布来降尺度图像。Dw构成按比例缩小的操作,具有正确的特定于图像的降级,随后将其用于训练数据集的生成。所得结果表明,我们的图像退化估计方法在提取内部图像统计信息以实现更好的超分辨率感知重建中的有效性。我们使用深度生成模型的功能来捕获测试图像中跨两个尺度的补丁的潜在表示,并训练降尺度的CNN Dw,以了解如何通过匹配这些潜在分布来降尺度图像。Dw构成按比例缩小的操作,具有正确的特定于图像的降级,随后将其用于训练数据集的生成。所得结果表明,我们的图像退化估计方法在提取内部图像统计信息以实现更好的超分辨率感知重建中的有效性。我们使用深度生成模型的功能来捕获测试图像中跨两个尺度的补丁的潜在表示,并训练降尺度的CNN Dw,以了解如何通过匹配这些潜在分布来降尺度图像。Dw构成按比例缩小的操作,具有正确的特定于图像的降级,随后将其用于训练数据集的生成。所得结果表明,我们的图像退化估计方法在提取内部图像统计信息以实现更好的超分辨率感知重建中的有效性。
更新日期:2021-05-22
down
wechat
bug