Tuning-Free Latent Diffusion Models for Ultrahigh-Resolution Image Editing
用于超高分辨率图像编辑的免调优潜在扩散模型
机构 * College of Computer Science and Artificial Intelligence, Wenzhou University(温州大学计算机科学与人工智能学院) ; Department of Computer Science, Memorial University of Newfoundland(纽芬兰纪念大学计算机科学系) ; AI Analytics Team, Nasdaq(纳斯达克人工智能分析团队) ; State Key Laboratory of CAD&CG, Zhejiang University(浙江大学CAD&CG国家重点实验室)
专题命中 扩散模型 :diffusion(title,abstract);image editing(title,abstract);image generation(abstract);分类 cs.CV、cs.MM
AI总结 针对现有图像编辑方法在高分辨率处理上的局限,提出免调优的UltraDiffEdit框架,通过多尺度渐进编辑、多补丁编码、全局-局部一致性去噪及补丁混合采样等技术,实现高质量超高分辨率图像编辑,处理能力达8K且灵活性高。
Comments 29 pages, 29 figures. Published in IEEE Transactions on Neural Networks and Learning Systems
Journal ref IEEE Transactions on Neural Networks and Learning Systems, 2026