X-Restormer++: 1st Place Solution for the UG2+ CVPR 2026 All-Weather Restoration Challenge

2026-05-13

AI Summary

This work addresses the challenge of image degradation under all-weather conditions by proposing an efficient restoration method built upon the X-Restormer framework. The approach introduces a spatially adaptive input scaling mechanism to accommodate multi-scale degradations across diverse weather scenarios, and incorporates a gradient-guided edge-aware (GGEA) loss function that synergistically combines multi-scale SSIM and L1 losses to significantly enhance structural fidelity and edge detail preservation. Furthermore, it integrates dual attention mechanisms—Multi-DConv Head Transposed Attention and Overlapping Cross-Attention—and is trained on a large-scale hybrid dataset. The proposed method achieved first place in the UG2+ CVPR 2026 All-Weather Image Restoration Challenge.

Abstract

In this work, we present our winning solution for the 8th UG2+ Challenge (CVPR 2026) Track 1: Image Restoration under All-weather Conditions. Our method is built upon the strong baseline framework X-Restormer, which effectively captures both channel-wise global dependencies and spatially-local structural information through its dual-attention design (Multi-DConv Head Transposed Attention and Overlapping Cross-Attention). To further boost the restoration performance, we propose several key improvements. First, we integrate the spatially-adaptive input scaling mechanism from Restormer-Plus to dynamically adjust the spatial weights of the input image, enhancing spatial adaptability. Second, to better preserve structural details and edge information, we introduce a novel Gradient-Guided Edge-Aware (GGEA) loss, which is combined with L1 and Multi-Scale SSIM losses in a unified training objective. Third, we significantly expand the training data by incorporating an extra 24,500 degraded-clean image pairs from FoundIR and WeatherBench alongside the original WeatherStream dataset. With these strategies, our proposed method successfully ranks the 1st place in the challenge.

Problem

• image restoration
• all-weather conditions
• degraded image
• weather degradation
• image enhancement

Innovation

• spatially-adaptive input scaling
• Gradient-Guided Edge-Aware loss
• dual-attention mechanism
• all-weather image restoration
• data augmentation