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

TL;DR

This paper presents X-Restormer++, a top-performing image restoration method for all-weather conditions, combining advanced attention mechanisms, adaptive input scaling, a new edge-aware loss, and expanded training data.

Contribution

It introduces several novel improvements to the X-Restormer framework, including spatially-adaptive input scaling, a gradient-guided edge-aware loss, and extensive data augmentation.

Findings

Achieved 1st place in the UG2+ Challenge CVPR 2026.

Enhanced restoration quality through dual-attention and adaptive scaling.

Improved structural detail preservation with GGEA loss.

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.