本研究致力于解决个性化图像增强问题，即根据用户偏好的图像为每位用户增强输入图像。以往方法对所有输入图像应用相同的偏好风格（即每位用户仅对应一种风格）；与这些方法不同，我们旨在通过考虑图像内容为每张图像应用不同风格，实现内容感知的个性化。为实现内容感知个性化，我们做出两项贡献。第一，提出一种名为masked style modeling（掩码风格建模）的方法，该方法借助masked language modeling（掩码语言建模）框架，能够结合内容预测输入图像的风格。第二，为使该模型能够考量图像内容，我们提出一种全新的训练方案：从Flickr下载图像，并使用退化模型生成伪输入-修图图像对。我们开展了定量评估与用户研究，采用该训练方案训练的方法成功实现了内容感知个性化；此外，我们的方法性能优于该领域其他现有方法。我们的源代码可在https://github.com/satoshi-kosugi/masked-style-modeling 获取。
英文摘要：
We address personalized image enhancement in this study, where we enhance input images for each user based on the user's preferred images. Previous methods apply the same preferred style to all input images (i.e., only one style for each user); in contrast to these methods, we aim to achieve content-aware personalization by applying different styles to each image considering the contents. For content-aware personalization, we make two contributions. First, we propose a method named masked style modeling, which can predict a style for an input image considering the contents by using the framework of masked language modeling. Second, to allow this model to consider the contents of images, we propose a novel training scheme where we download images from Flickr and create pseudo input and retouched image pairs using a degrading model. We conduct quantitative evaluations and a user study, and our method trained using our training scheme successfully achieves content-aware personalization; moreover, our method outperforms other previous methods in this field. Our source code is available at https://github.com/satoshi-kosugi/masked-style-modeling.
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；School of Future Technology, University of Chinese Academy of Sciences（中国科学院大学未来技术学院）
Self-supervised blind denoising for Poisson-Gaussian noise remains a challenging task. Pseudo-supervised pairs constructed from single noisy images re-corrupt the signal and degrade the performance. The visible blindspots solve the information loss in masked inputs. However, without explicitly noise sensing, mean square error as an objective function cannot adjust denoising intensities for dynamic noise levels, leading to noticeable residual noise. In this paper, we propose Blind2Sound, a simple yet effective approach to overcome residual noise in denoised images. The proposed adaptive re-visi…
Harnessing Massive Satellite Imagery with Efficient Masked Image Modeling
College of Computer Science and Technology, National University of Defense Technology（计算机科学与技术学院，国防科技大学）；Tsinghua University（清华大学）；School of Computer Science, Wuhan University（计算机科学学院，武汉大学）；Zhongguancun Academy（中关村学院）；Nankai University（南开大学）
LumiCtrl : Learning Illuminant Prompts for Lighting Control in Personalized Text-to-Image Models
Computer Vision Center, Spain（西班牙计算机视觉中心）；City University of Hong Kong（香港城市大学）；Computer Sciences Department, Universitat Autònoma de Barcelona, Spain（西班牙巴塞罗那自治大学计算机科学系）；Program of Computer Science, City University of Hong Kong (Dongguan)（香港城市大学（东莞）计算机科学项目）
A Survey on Personalized Content Synthesis with Diffusion Models
Department of Computing（计算系）；The Hong Kong Polytechnic University（香港理工大学）；Center for Artificial Intelligence and Robotics（人工智能与机器人中心）；Hong Kong Institute of Science & Innovation（香港科学创新研究院）；Chinese Academy of Sciences（中国科学院）；State Key Laboratory of Multimodal Artificial Intelligence Systems（多模态人工智能系统国家重点实验室）；Chinese Academy of Sciences Institute of Automation（中国科学院自动化研究所）；School of Artificial Intelligence（人工智能学院）；University of Chinese Academy of Sciences（中国科学院大学）
