研究介绍
3D 高斯投影（3D Gaussian Splatting，简称 3DGS）在辐射场重建方面取得了突破，实现了高效且高保真的新视角合成。然而，由于 3DGS 的非结构化特性，在大规模和复杂场景中准确表示表面仍然是一个重大挑战。本文提出了 CityGaussianV2，这是一种面向大规模场景重建的新方法，旨在解决几何精度和效率方面的关键问题。
该方法基于 2D 高斯投影（2DGS）良好的泛化能力，着重解决其收敛性与可扩展性问题。具体而言，我们引入了基于梯度分解的密化与深度回归技术，以消除模糊伪影并加速收敛过程。为了解决扩展性问题，我们设计了一种拉伸滤波器，以缓解因 2DGS 退化引起的高斯数量爆炸问题。
此外，我们对 CityGaussian 管线进行了并行训练优化，实现了高达 10 倍的数据压缩，训练时间节省至少 25%，内存使用减少 50%。我们还建立了大规模场景下的标准几何评测基准。实验结果表明，我们的方法在视觉质量、几何精度以及存储与训练成本之间取得了良好的平衡。
更多实时演示和官方代码实现可见项目主页：https://dekuliutesla.github.io/CityGaussianV2
Abstract
Recently, 3D Gaussian Splatting (3DGS) has revolutionized radiance field reconstruction, manifesting efficient and high-fidelity novel view synthesis. However,
accurately representing surfaces, especially in large and complex scenarios, remains a significant challenge due to the unstructured nature of 3DGS. In this paper,
we present CityGaussianV2, a novel approach for large-scale scene reconstruction
that addresses critical challenges related to geometric accuracy and efficiency.
Building on the favorable generalization capabilities of 2D Gaussian Splatting
(2DGS), we address its convergence and scalability issues. Specifically, we implement a decomposed-gradient-based densification and depth regression technique
to eliminate blurry artifacts and accelerate convergence. To scale up, we introduce
an elongation filter that mitigates Gaussian count explosion caused by 2DGS degeneration. Furthermore, we optimize the CityGaussian pipeline for parallel training, achieving up to 10× compression, at least 25% savings in training time, and
a 50% decrease in memory usage. We also established standard geometry benchmarks under large-scale scenes. Experimental results demonstrate that our method
strikes a promising balance between visual quality, geometric accuracy, as well as
storage and training costs. More live demos and official code implementation are
available at our project page: https://dekuliutesla.github.io/CityGaussianV2
