The primary aim of single-image super-resolution is to construct high-resolution (HR) images from corresponding low-resolution (LR) inputs. In previous approaches, which have generally been supervised, the training objective typically measures a pixel-wise average distance between the super-resolved (SR) and HR images. Optimizing such metrics often leads to blurring, especially in high variance (detailed) regions. We propose an alternative formulation of the super-resolution problem based on creating realistic SR images that downscale correctly. We present an algorithm addressing this problem, PULSE (Photo Upsampling via Latent Space Exploration), which generates high-resolution, realistic images at resolutions previously unseen in the literature. It accomplishes this in an entirely self-supervised fashion and is not confined to a specific degradation operator used during training, unlike previous methods (which require supervised training on databases of LR-HR image pairs). Instead of starting with the LR image and slowly adding detail, PULSE traverses the high-resolution natural image manifold, searching for images that downscale to the original LR image. This is formalized through the "downscaling loss," which guides exploration through the latent space of a generative model. By leveraging properties of high-dimensional Gaussians, we restrict the search space to guarantee realistic outputs. PULSE thereby generates super-resolved images that both are realistic and downscale correctly. We show proof of concept of our approach in the domain of face super-resolution (i.e., face hallucination). We also present a discussion of the limitations and biases of the method as currently implemented with an accompanying model card with relevant metrics. Our method outperforms state-of-the-art methods in perceptual quality at higher resolutions and scale factors than previously possible.
Single-sample image-fusion upsampling of fluorescence lifetime images
School of Physics & Astronomy, University of Glasgow（格拉斯哥大学物理与天文学院）；Center for Physical Sciences and Technology（物理科学与技术中心）；Cancer Research UK, Beatson Institute（英国癌症研究基金会Beatson研究所）；School of Health & Life Sciences, University of the West of Scotland（西苏格兰大学健康与生命科学学院）；University College London（伦敦大学学院）；University of Cambridge（剑桥大学）；Department of Laser Technologies（激光技术系）
Universität Innsbruck（因斯布鲁克大学）；Universität Graz（格拉茨大学）；Research Center for Non Destructive Testing GmbH（非破坏性测试研究中心 GmbH）；Prospective Instruments LK OG（前瞻性仪器 LK OG）
Technical University of Munich（慕尼黑工业大学）；Advanced Digital Sciences Center, University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校高级数字科学中心）；GREYC laboratory, UMR CNRS 6072（法国国家科学研究中心6072联合研究单元GREYC实验室）
High-Resolution Daytime Translation Without Domain Labels
Samsung AI Center, Moscow（三星人工智能中心（莫斯科））；National Research University Higher School of Economics, St.-Petersburg（国立研究型大学高等经济学院（圣彼得堡））；Skolkovo Institute of Science and Technology, Moscow（斯科尔科沃科学技术研究院（莫斯科））
