Self-supervised Dynamic Heterogeneous Degradation Modeling for Unified Zero-Shot Image Restoration 事件

Self-supervised Dynamic Heterogeneous Degradation Modeling for Unified Zero-Shot Image Restoration arXiv:2605.24593v1 Announce Type: new Abstract: Zero-shot image restoration provides a flexible way to handle diverse degradations without task-specific training. However, existing methods typically rely on stacked layers or pre-trained features to enhance degradation expression, while overlooking physically consistent priors. The insufficient degradation prompts impose the heavy training burden a