arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.00953eess.IVcs.CV
收藏

用于真实世界超分辨率的深度生成对抗残差卷积网络

Deep Generative Adversarial Residual Convolutional Networks for Real-World Super-Resolution

University of Udine(乌迪内大学)

Rao Muhammad Umer, Gian Luca Foresti, Christian Micheloni

摘要:

目前大多数基于深度学习的单图像超分辨率(SISR)方法都专注于设计更深/更宽的模型,以从大量配对(LR/HR)训练数据中学习低分辨率(LR)输入和高分辨率(HR)输出之间的非线性映射。他们通常假设LR图像是HR图像的双三次下采样版本。然而,这样的降级过程在现实世界的设置中是不可用的,即固有的传感器噪声、随机噪声、压缩伪影、图像降级过程与相机设备之间的可能失配。由于真实世界的图像损坏,它显着降低了当前SISR方法的性能。为了解决这些问题,我们提出了一个深度超分辨率残差卷积生成对抗网络(SRResCGAN),通过对抗训练模型,在HR域中从其生成的LR对应物中进行像素监督,来跟踪真实世界的退化设置。该网络通过使用强大的图像正则化和凸优化技术最小化基于能量的目标函数来利用残差学习。我们在定量和定性实验中证明了我们提出的方法,这些实验可以鲁棒地推广到真实输入,并且很容易部署到其他降尺度运营商和移动/嵌入式设备。

英文摘要:

Most current deep learning based single image super-resolution (SISR) methods focus on designing deeper / wider models to learn the non-linear mapping between low-resolution (LR) inputs and the high-resolution (HR) outputs from a large number of paired (LR/HR) training data. They usually take as assumption that the LR image is a bicubic down-sampled version of the HR image. However, such degradation process is not available in real-world settings i.e. inherent sensor noise, stochastic noise, compression artifacts, possible mismatch between image degradation process and camera device. It reduces significantly the performance of current SISR methods due to real-world image corruptions. To address these problems, we propose a deep Super-Resolution Residual Convolutional Generative Adversarial Network (SRResCGAN) to follow the real-world degradation settings by adversarial training the model with pixel-wise supervision in the HR domain from its generated LR counterpart. The proposed network exploits the residual learning by minimizing the energy-based objective function with powerful image regularization and convex optimization techniques. We demonstrate our proposed approach in quantitative and qualitative experiments that generalize robustly to real input and it is easy to deploy for other down-scaling operators and mobile/embedded devices.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑