arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1904.00523cs.CV
收藏

面向真实世界的单幅图像超分辨率:一种新的基准和新的模型

Toward Real-World Single Image Super-Resolution: A New Benchmark and A New Model

The Hong Kong Polytechnic University(香港理工大学);DJI Co.,Ltd(大疆创新科技有限公司);DAMO Academy, Alibaba Group(阿里巴巴集团达摩院)

Jianrui Cai, Hui Zeng, Hongwei Yong, Zisheng Cao, Lei Zhang

摘要:

大多数现有的基于学习的单图像超分辨率(SISR)方法都是在模拟数据集上训练和评估的,其中低分辨率(LR)图像是通过应用简单且均匀的退化(即,双三次下采样)到它们的高分辨率(HR)对应物。然而,现实世界LR图像中的退化要复杂得多。因此,在模拟数据上训练的SISR模型在应用于实际场景时变得不那么有效。在本文中,我们建立了一个真实世界的超分辨率(RealSR)数据集,在同一场景的配对LR-HR图像通过调整焦距的数码相机捕获。提出了一种图像配准算法,用于逐步对准不同分辨率的图像对。考虑到退化核在我们的数据集中自然是不均匀的,我们提出了一个基于拉普拉斯金字塔的核预测网络(LP-KPN),它有效地学习每像素的核来恢复HR图像。我们广泛的实验表明,在我们的RealSR数据集上训练的SISR模型比在模拟数据集上训练的模型在真实场景中提供更好的视觉质量,具有更清晰的边缘和更精细的纹理。虽然我们的RealSR数据集仅使用两台相机(Canon 5D3和Nikon D810)构建,但经过训练的模型可以很好地推广到其他相机设备,如索尼a7 II和手机。

英文摘要:

Most of the existing learning-based single image superresolution (SISR) methods are trained and evaluated on simulated datasets, where the low-resolution (LR) images are generated by applying a simple and uniform degradation (i.e., bicubic downsampling) to their high-resolution (HR) counterparts. However, the degradations in real-world LR images are far more complicated. As a consequence, the SISR models trained on simulated data become less effective when applied to practical scenarios. In this paper, we build a real-world super-resolution (RealSR) dataset where paired LR-HR images on the same scene are captured by adjusting the focal length of a digital camera. An image registration algorithm is developed to progressively align the image pairs at different resolutions. Considering that the degradation kernels are naturally non-uniform in our dataset, we present a Laplacian pyramid based kernel prediction network (LP-KPN), which efficiently learns per-pixel kernels to recover the HR image. Our extensive experiments demonstrate that SISR models trained on our RealSR dataset deliver better visual quality with sharper edges and finer textures on real-world scenes than those trained on simulated datasets. Though our RealSR dataset is built by using only two cameras (Canon 5D3 and Nikon D810), the trained model generalizes well to other camera devices such as Sony a7II and mobile phones.

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。

↑