Video denoising is fundamental to low-level vision and real-world imaging, yet existing self-supervised methods remain fragile under severe noise and complex motion. Most approaches still rely on spatially and temporally discrete grid-based representations: bl... ...

请注册登录后继续浏览