先去烟再三维重建，用物理先验提升新视角合成质量。
Dehaze-then-Splat: Generative Dehazing with Physics-Informed 3D Gaussian Splatting for Smoke-Free Novel View Synthesis
- 分两阶段：先生成无烟图像，再用物理约束训练3D高斯点云
- 新视角合成达20.98dB PSNR，比基线提升1.5dB
- 适合需要高质量3D重建的烟雾场景应用
我们提出Dehaze-then-Splat，一种用于NTIRE 2026 3D修复与重建挑战赛第2赛道的多视角去烟与新视角合成两阶段方法。第一阶段通过Nano Banana Pro对每帧进行生成式去烟，并进行亮度归一化；第二阶段训练3D高斯点云（3DGS），引入物理启发的辅助损失：基于伪深度的皮尔逊相关性深度监督、暗通道先验正则化以及双源梯度匹配，以弥补逐帧生成处理带来的跨视图不一致性。我们发现去烟后重建中存在根本矛盾：单帧去烟质量高并不保证多视角一致性，后者会导致渲染模糊和结构不稳定。分析表明，结合早期停止的马尔可夫链蒙特卡洛稠密化与深度及去雾先验，能有效缓解此类伪影。在Akikaze验证场景中，该方法的新视角合成达到20.98dB PSNR和0.683 SSIM，较未正则化基线提升1.50dB。
原文摘要 · Abstract (English)
We present Dehaze-then-Splat, a two-stage pipeline for multi-view smoke removal and novel view synthesis developed for Track~2 of the NTIRE 2026 3D Restoration and Reconstruction Challenge. In the first stage, we produce pseudo-clean training images via per-frame generative dehazing using Nano Banana Pro, followed by brightness normalization. In the second stage, we train 3D Gaussian Splatting (3DGS) with physics-informed auxiliary losses -- depth supervision via Pearson correlation with pseudo-depth, dark channel prior regularization, and dual-source gradient matching -- that compensate for cross-view inconsistencies inherent in frame-wise generative processing. We identify a fundamental tension in dehaze-then-reconstruct pipelines: per-image restoration quality does not guarantee multi-view consistency, and such inconsistency manifests as blurred renders and structural instability in downstream 3D reconstruction.Our analysis shows that MCMC-based densification with early stopping, combined with depth and haze-suppression priors, effectively mitigates these artifacts. On the Akikaze validation scene, our pipeline achieves 20.98\,dB PSNR and 0.683 SSIM for novel view synthesis, a +1.50\,dB improvement over the unregularized baseline.
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。