ICF-SRSR：用于自监督真实世界单图像超分辨率的可逆尺度条件函数
ICF-SRSR: Invertible scale-Conditional Function for Self-Supervised Real-world Single Image Super-Resolution
AI总结：
提出可逆尺度条件函数（ICF）构建自监督SISR框架（ICF-SRSR），无需训练数据即可处理真实世界超分辨率，并生成LR-HR对增强现有方法，性能优于合成训练方法且与最先进方法相当。
摘要：
单图像超分辨率（SISR）是一个具有挑战性的不适定问题，旨在将给定的低分辨率（LR）图像上采样到高分辨率（HR）对应图像。由于难以获得真实的LR-HR训练对，近期的方法通常在由简化下采样算子（如双三次插值）模拟的LR图像上进行训练。这种方法在实践中可能存在问题，因为合成LR图像与真实世界LR图像之间存在较大差距。为缓解这一问题，我们提出了一种新颖的可逆尺度条件函数（ICF），该函数能够缩放输入图像，并在不同尺度条件下恢复原始输入。通过利用所提出的ICF，我们构建了一个新颖的自监督SISR框架（ICF-SRSR），以在无需任何成对/非成对训练数据的情况下处理真实世界的SR任务。此外，我们的ICF-SRSR能够生成真实且可行的LR-HR对，这可以使现有的监督式SISR网络更加鲁棒。大量实验证明了所提方法在全自监督方式下处理SISR的有效性。在真实世界场景中，我们的ICF-SRSR相比在合成配对图像上训练的现有方法展现出更优越的性能，并且在公共基准数据集上与最先进的监督/非监督方法相比表现出相当的性能。
英文摘要：
Single image super-resolution (SISR) is a challenging ill-posed problem that aims to up-sample a given low-resolution (LR) image to a high-resolution (HR) counterpart. Due to the difficulty in obtaining real LR-HR training pairs, recent approaches are trained on simulated LR images degraded by simplified down-sampling operators, e.g., bicubic. Such an approach can be problematic in practice because of the large gap between the synthesized and real-world LR images. To alleviate the issue, we propose a novel Invertible scale-Conditional Function (ICF), which can scale an input image and then restore the original input with different scale conditions. By leveraging the proposed ICF, we construct a novel self-supervised SISR framework (ICF-SRSR) to handle the real-world SR task without using any paired/unpaired training data. Furthermore, our ICF-SRSR can generate realistic and feasible LR-HR pairs, which can make existing supervised SISR networks more robust. Extensive experiments demonstrate the effectiveness of the proposed method in handling SISR in a fully self-supervised manner. Our ICF-SRSR demonstrates superior performance compared to the existing methods trained on synthetic paired images in real-world scenarios and exhibits comparable performance compared to state-of-the-art supervised/unsupervised methods on public benchmark datasets.
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。
