用物理先验生成伪清晰监督，实现烟雾场景的高质量3D重建。
SmokeGS-R: Physics-Guided Pseudo-Clean 3DGS for Real-World Multi-View Smoke Restoration
- 分两步走：先重建几何，再用物理模型修正外观。
- 在真实烟雾数据上达到PSNR=15.21、SSIM=0.666，优于基线3.68dB。
- 适合做烟雾环境下的3D视觉重建与渲染任务的研究者。
真实世界烟雾同时衰减场景辐射、引入大气光，并破坏多视角外观一致性，导致鲁棒3D重建极为困难。本文提出SmokeGS-R，针对NTIRE 2026 3D恢复与重建挑战赛第2赛道设计的实用流程。核心思路是将几何恢复与外观校正解耦：通过优化的暗通道先验和引导滤波生成物理引导的伪清晰监督，训练一个仅关注清晰外观的3D高斯点云源模型，并通过几何均值参考聚合、LAB空间Reinhard转移和轻量高斯平滑，将其渲染结果与捐赠模型集进行调和。在官方测试排行榜上，最终提交结果达PSNR=15.217、SSIM=0.666。公开发布RealX3D后，对同一冻结结果在七个释放场景上重评估，未重新训练即获PSNR=15.209、SSIM=0.644、LPIPS=0.551，优于同场景最强官方基线平均值3.68dB。结果表明，以几何优先重建结合稳定后期外观调和，是真实多视角烟雾恢复的有效方案。代码已开源：https://github.com/windrise/3drr_Track2_SmokeGS-R。
原文摘要 · Abstract (English)
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 https://github.com/windrise/3drr_Track2_SmokeGS-R.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。