TPGDiff : Hierarchical Triple-Prior Guided Diffusion for Image Restoration

All-in-one image restoration aims to address diverse degradation types using a single unified model. Existing methods typically rely on degradation priors to guide restoration, yet often struggle to reconstruct content in severely degraded regions. Although recent works leverage semantic information to facilitate content generation, integrating it into the shallow layers of diffusion models often disrupts spatial structures (e.g., blurring artifacts). To address this issue, we propose a Triple-Prior Guided Diffusion (TPGDiff) network for unified image restoration. TPGDiff incorporates degradation priors throughout the diffusion trajectory, while introducing structural priors into shallow layers and semantic priors into deep layers, enabling hierarchical and complementary prior guidance for image reconstruction. Specifically, we leverage multi-source structural cues as structural priors to capture fine-grained details and guide shallow layers representations. To complement this design, we further develop a distillation-driven semantic extractor that yields robust semantic priors, ensuring reliable high-level guidance at deep layers even under severe degradations. Furthermore, a degradation extractor is employed to learn degradation-aware priors, enabling stage-adaptive control of the diffusion process across all timesteps. Extensive experiments on both single- and multi-degradation benchmarks demonstrate that TPGDiff achieves superior performance and generalization across diverse restoration scenarios.

All-in-one image restoration aims to handle diverse degradations with a unified model, but existing methods often struggle to recover reliable content in severely degraded regions. While semantic information has been introduced to enhance content generation, injecting it into shallow diffusion layers may disrupt spatial structures and cause blurring artifacts. To address this problem, we propose Triple-Prior Guided Diffusion (TPGDiff), a unified restoration framework that hierarchically integrates degradation, structural, and semantic priors. Specifically, degradation priors are incorporated throughout the diffusion trajectory to enable stage-adaptive restoration, structural priors are introduced into shallow layers to preserve fine-grained details and spatial layouts, and robust semantic priors are distilled to guide deep layers for high-level content reconstruction. Extensive experiments on single- and multi-degradation benchmarks demonstrate that TPGDiff achieves superior restoration performance and strong generalization across diverse degradation scenarios.