用自适应非线性曲线提升暗光3D重建的伪真值质量
Scene-Adaptive Nonlinear Tone Curves for Pseudo Ground-Truth Generation in Low-Light 3D Gaussian Splatting
- 设计自适应非线性色调曲线替代传统线性增强，动态优化暗区亮度
- 在LOM和RealX3D数据集上分别实现最高+4.34dB和+3.25dB的PSNR提升
- 无需修改3DGS主干，适配各类低光3D重建场景
暗光新视角合成因多视角图像噪声大、结构细节弱、动态范围压缩而困难。近期3D高斯泼溅（3DGS）方法在缺乏正常光照参考时，通过生成伪真值（pseudo-GT）图像作为监督目标来应对。现有伪真值方法对所有像素采用统一线性增益，导致亮区过曝、暗区增强不足，限制重建质量。我们发现，2D暗光增强中长期使用的非线性色调映射尚未被用于3D重建的伪真值生成。为此，提出一种场景自适应非线性色调曲线框架，替换原有线性伪真值。该框架引入基于百分位数的归一化实现无场景依赖曲线应用，结合场景自适应偏移自动调整黑电平，并设计两种互补曲线：有界指数型自适应SoftExp（ASE）与数据驱动型三次多项式自适应Poly3（AP3）。模块仅改变伪真值计算，不改动3DGS主干。在涵盖21个场景的三个基准测试中，两种曲线均显著优于线性基线，在LOM上最高提升+4.34dB，RealX3D上+3.25dB。尽管数学形式不同，两者性能相当，表明提升效果具有曲线无关性。代码已开源。
原文摘要 · Abstract (English)
Low-light novel view synthesis is challenging because dark multi-view images contain noise, weak structural detail, and compressed dynamic range. Recent 3D Gaussian Splatting (3DGS) methods address these challenges by generating pseudo ground-truth (pseudo-GT) images as supervision targets when paired normal-light references are unavailable. Existing pseudo-GT methods apply a uniform linear gain to all pixels, which clips bright regions while providing insufficient enhancement in dark regions, limiting reconstruction quality. We observe that nonlinear tone mappings, long established in 2D low-light enhancement, have not been explored for pseudo-GT generation in 3D reconstruction. Accordingly, we propose a scene-adaptive nonlinear tone-curve framework that replaces linear pseudo-GT with nonlinear alternatives. The framework introduces percentile-based normalisation for scene-agnostic curve application, a scene-adaptive offset for automatic black-level adjustment, and two complementary curves: Adaptive SoftExp (ASE), a bounded exponential curve, and Adaptive Poly3 (AP3), a data-driven cubic polynomial. The module changes only the pseudo-GT computation and leaves the 3DGS backbone unchanged. Experiments on three benchmarks covering 21 scenes show that both curves consistently outperform the linear baseline with PSNR improvements up to +4.34 dB on LOM and +3.25 dB on RealX3D. Both curves achieve similar performance despite their different mathematical forms, suggesting the improvement is curve-agnostic. Code is available at https://github.com/lvmingzhe/adaptiveToneCurve
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。