NGPS: Structure-Preserving Self-Supervised Denoising via Neighbor-Guided Patch Sampling
NGPS:基于邻域引导的补丁采样的结构保持自监督去噪
机构 * Department of Artificial Intelligence, Chung-Ang University(Chung-Ang大学人工智能系) ; SNUAILAB
AI总结 提出NGPS框架,通过邻域引导的补丁采样在局部错位下构建自监督去噪目标,无需显式配准,在CT和MRI上提升保真度和结构敏感指标。
Comments The 19th European Conference on Computer Vision: ECCV 2026