Abstract
Real-world smoke simultaneously attenuates scene radiance, adds airlight, and destabilizes multi-view appearance consistency, making robust 3D reconstruction particularly difficult. We present \textbf{SmokeGS-R}, a practical pipeline developed for the NTIRE 2026 3D Restoration and Reconstruction Track 2 challenge. The key idea is to decouple geometry recovery from appearance correction: we generate physics-guided pseudo-clean supervision with a refined dark channel prior and guided filtering, train a sharp clean-only 3D Gaussian Splatting source model, and then harmonize its renderings with a donor ensemble using geometric-mean reference aggregation, LAB-space Reinhard transfer, and light Gaussian smoothing. On the official challenge testing leaderboard, the final submission achieved \mbox{PSNR $=15.217$} and \mbox{SSIM $=0.666$}. After the public release of RealX3D, we re-evaluated the same frozen result on the seven released challenge scenes without retraining and obtained \mbox{PSNR $=15.209$}, \mbox{SSIM $=0.644$}, and \mbox{LPIPS $=0.551$}, outperforming the strongest official baseline average on the same scenes by $+3.68$ dB PSNR. These results suggest that a geometry-first reconstruction strategy combined with stable post-render appearance harmonization is an effective recipe for real-world multi-view smoke restoration. The code is available at this https URL.
Abstract (translated)
真实世界中的烟雾会同时衰减场景辐射度、增加大气光并破坏多视角外观一致性，使得鲁棒的3D重建尤为困难。我们提出了\textbf{SmokeGS-R}，一个为NTIRE 2026 3D Restoration and Reconstruction Track 2挑战赛开发的实用流程。核心思想是将几何恢复与外观校正解耦：我们通过改进的暗通道先验和引导滤波生成物理引导的伪清洁监督数据，训练一个锐利的纯清洁3D高斯泼溅源模型，然后利用几何平均参考聚合、LAB空间Reinhard传递和轻量高斯平滑，通过捐赠者集成协调其渲染结果。在官方挑战测试排行榜上，最终提交结果达到了\mbox{PSNR $=15.217$}和\mbox{SSIM $=0.666$}。在RealX3D公开发布后，我们在未重新训练的情况下，对已发布的七个挑战场景重新评估了同一冻结结果，获得了\mbox{PSNR $=15.209$}、\mbox{SSIM $=0.644$}和\mbox{LPIPS $=0.551$}，在同一场景上比最强官方基线平均提升了$+3.68$ dB PSNR。这些结果表明，结合几何优先的重建策略与稳定的后渲染外观协调，是处理真实世界多视角烟雾恢复的有效方案。代码可在以下链接获取：this https URL。
URL
https://arxiv.org/abs/2604.05301
