发表机构
TU Braunschweig; Università della Svizzera italiana; University of Würzburg; University of New Mexico（布伦瑞克工业大学; 意大利南部大学; 维尔茨堡大学; 新墨西哥大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究对比了SH与近期球外观模型，提出基于小型共享MLP的紧凑神经外观模型，将3D高斯溅射每个基元外观占用空间从192字节减至28字节，优化提速1.3倍且提升重建质量，为替换SH提供实用指导。
AI 中文摘要
基于显式基元的辐射场，如3D高斯溅射，通常使用低阶球谐函数（SH）建模视图相关外观。尽管SH评估高效，但其系数主导每个基元的存储和内存流量，且带限基函数限制了角度细节。我们对SH与近期球外观模型进行了全面端到端比较，并引入一种隐式替代方案，该方案使用小型共享多层感知器（MLP）解码紧凑的每个基元潜在代码。我们将所有模型集成到同一优化流水线中，将其前向和反向传播融合为可微CUDA光栅化器，并提供适用于笔记本电脑和移动GPU的便携式WebGL查看器。我们在重建质量、内存使用以及优化和渲染性能方面的评估表明，近期球外观模型提供了最强的整体质量-效率权衡。我们的神经表示是所评估模型中最紧凑的，与三阶SH相比，它将每个基元的外观占用空间从192字节减少到28字节，使优化速度提高1.3倍，同时提升了重建质量。我们进一步分析了外观参数化如何影响优化，确定了恢复几何的差异，以及表达性模型吸收非静态场景内容的趋势。总之，我们的框架和分析为替换SH提供了超出仅图像指标所能捕捉的实用指导。
英文摘要
Explicit primitive-based radiance fields such as 3D Gaussian Splatting typically model view-dependent appearance using low-order spherical harmonics (SH). Although efficient to evaluate, SH coefficients dominate per-primitive storage and memory traffic, while their band-limited basis restricts angular detail. We present a thorough, end-to-end comparison of SH and recent spherical appearance models and introduce an implicit alternative that decodes compact per-primitive latent codes using a tiny shared MLP. We integrate all models into the same optimized pipeline, fusing their forward and backward passes into a differentiable CUDA rasterizer and provide a portable WebGL viewer for laptop and mobile GPUs. Our evaluation across reconstruction quality, memory use, and optimization and rendering performance shows that recent spherical models offer the strongest overall quality-efficiency trade-off. Our neural representation is the most compact model evaluated and, compared to third-degree SH, reduces the per-primitive appearance footprint from 192 to 28 bytes, accelerates optimization by 1.3$\times$, while improving reconstruction quality. We further analyze how appearance parametrization shapes optimization, identifying differences in recovered geometry and the tendency of expressive models to absorb non-static scene content. Together, our framework and analysis provide practical guidance for replacing SH beyond what image metrics alone can capture.
CommentsProject page: https://fhahlbohm.github.io/efficient-gaussian-appearance
