GraLoD, a plug-and-play framework that treats restoration scale as a spatially varying and stage-dependent continuous variable, is proposed and minimal-sufficient footprint calibration (MSFC) together with structure-aware regularization (SAR) is introduced to encourage restoration-effective and spatially coherent LOD assignments.
Abstract
The spatial support required for image restoration varies across degradation types, image regions, and reconstruction stages. However, most existing methods rely on predefined multi-scale hierarchies and aggregate features through fixed fusion or attention, leaving the representation scale itself largely determined by the network architecture. This limitation becomes more pronounced when a task-specific backbone is extended to heterogeneous degradations in all-in-one restoration. Inspired by level-of-detail (LOD) rendering in computer graphics, we propose GraLoD, a plug-and-play framework that treats restoration scale as a spatially varying and stage-dependent continuous variable. GraLoD reuses the native encoder hierarchy, aligns its multi-scale features into a shared LOD representation space, and predicts a stage-conditioned LOD field at each decoder stage. Each spatial location then continuously queries only two neighboring representation levels, enabling the effective restoration scale to adapt to both local image content and reconstruction progress. To prevent degenerate or arbitrary scale selection, we further introduce minimal-sufficient footprint calibration (MSFC) together with structure-aware regularization (SAR) to encourage restoration-effective and spatially coherent LOD assignments. GraLoD can be directly integrated into existing restoration backbones without redesigning their fundamental feature-processing blocks. Extensive experiments demonstrate consistent improvements in task-specific and all-in-one restoration.
Real images often exhibit unknown and mixed degradations, making restoration substantially more challenging than single-task image restoration because multiple distortion types interact within the same observation. Consequently, existing methods often rely on prior knowledge of the degradation type or separate task-spe...
T. Nadeem, Arslan Majal, Muhammad Tahir et al.· IEEE Transactions on Instrum...· 0 citations
MagnifiQ is introduced, an image restoration framework that progressively upscales and restores images across resolutions, e.g., from 1024x1024 to 4096x4096, and leverages a pre-trained text-to-image diffusion model such as SDXL and adapts it for more scalable high-resolution inference by replacing its original self-at...
M. Reddy, Yashesh Savani, Antoine Mercier et al.· 0 citations
Recent diffusion-based pipelines have achieved promising progress in image-to-3D synthesis. However, generating high-fidelity details remains challenging, especially when the input image contains rich details. Existing approaches often rely on globally encoded conditioning features, which compress spatial information a...
Jun-Yu Li, Qiu-Yu Chen, Peng-Cheng Wang et al.· 0 citations
An adaptive multi-scale decoding framework that effectively balances global context with fine-grained detail is proposed that exhibits superior robustness and generalization across diverse domains, effectively alleviating limitations of existing fusion-based approaches.
This paper develops a Retinex-guided state space backbone that integrates illumination–reflectance priors into the network in a structured manner, enabling more reliable restoration of shadow regions, and introduces a Shadow Interaction Module within a multi-scale channel attention framework.
Wei-Ju Zhao, Run-Xing Zhao, Shi-Jian Yang· Signal, Image and Video Proc...· 0 citations
PixRestore is presented, a VAE-free pixel-space Diffusion Transformer (DiT) for UIR, where the diffusion backbone is trained entirely from scratch, without relying on T2I pretraining.
Ling-Chen Sun, Rong-Yuan Wu, Xiang-Tao Kong et al.· 2 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.