用立体视觉先验提升稀疏视角3D高斯点云的几何精度
StereoGS: Sparse-View 3D Gaussian Splatting via Stereo Priors
- 通过构建虚拟立体对并利用基础立体模型，实现绝对尺度与双目一致性约束
- 在LLFF等数据集上达到当前最优稀疏视图重建效果，且推理无额外开销
- 适合需要高精度3D重建的稀疏视角场景，如自动驾驶、AR/VR应用
3D高斯点云（3DGS）在实时新视角合成中表现优异，但在稀疏视角下因几何约束不足易出现过拟合。现有方法引入单目深度先验，但存在尺度模糊和跨视角不一致问题，导致几何缺陷。本文提出StereoGS，一种融合立体先验的稀疏视角3DGS框架，通过优化过程中构建虚拟立体对，并利用基础立体模型强制绝对尺度与双目一致性结构。为抑制过拟合并消除冗余高斯点，设计基于梯度感知的透明度衰减策略，动态惩罚相对透明度梯度较大的高斯点。结合零样本多视角深度估计的致密初始化策略，有效将高斯点锚定在准确场景表面。在LLFF、DTU、Mip-NeRF360和Blender数据集上的大量实验表明，StereoGS在稀疏视角设置下取得当前最优性能，且不增加任何推理开销。
原文摘要 · Abstract (English)
3D Gaussian Splatting (3DGS) has achieved remarkable success in real-time novel view synthesis, yet it suffers from severe overfitting under sparse-view settings due to insufficient geometric constraints. While recent methods introduce monocular depth priors to mitigate this, they inherently struggle with scale ambiguity and cross-view inconsistency, leading to defective geometry. In this paper, we propose StereoGS, a novel sparse-view 3DGS framework that integrates stereo priors to establish reliable binocular consistency. Unlike scale-agnostic monocular constraints, StereoGS introduces a Stereo Depth Regularization by constructing virtual stereo pairs during optimization and leveraging a foundation stereo model to enforce absolute scale and binocular-consistent structures. To further suppress overfitting and eliminate redundant primitives, we design a Gradient-Aware Opacity Decay strategy that dynamically penalizes Gaussians based on their relative opacity gradient magnitudes. Combined with a Consistency-Aware Dense Initialization using zero-shot multi-view depth estimation, StereoGS effectively anchors primitives to accurate scene surfaces. Extensive experiments on LLFF, DTU, Mip-NeRF360, and Blender datasets demonstrate that StereoGS achieves state-of-the-art performance in sparse-view settings without incurring any additional inference overhead. Project Page: https://stringerywh00.github.io/StereoGS_project_page/
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。