We present MVSGaussian, a new generalizable 3D Gaussian representation approach derived from Multi-View Stereo (MVS) that can efficiently reconstruct unseen scenes. Specifically, 1) we leverage MVS to encode geometry-aware Gaussian representations and decode them into Gaussian parameters. 2) To further enhance performance, we propose a hybrid Gaussian rendering that integrates an efficient volume rendering design for novel view synthesis. 3) To support fast fine-tuning for specific scenes, we introduce a multi-view geometric consistent aggregation strategy to effectively aggregate the point clouds generated by the generalizable model, serving as the initialization for per-scene optimization. Compared with previous generalizable NeRF-based methods, which typically require minutes of fine-tuning and seconds of rendering per image, MVSGaussian achieves real-time rendering with better synthesis quality for each scene. Compared with the vanilla 3D-GS, MVSGaussian achieves better view synthesis with less training computational cost. Extensive experiments on DTU, Real Forward-facing, NeRF Synthetic, and Tanks and Temples datasets validate that MVSGaussian attains state-of-the-art performance with convincing generalizability, real-time rendering speed, and fast per-scene optimization.

翻译：我们提出了MVSGaussian，一种源自多视图立体（MVS）的新型可泛化3D高斯表示方法，能够高效重建未见场景。具体而言：1）利用MVS编码几何感知的高斯表示，并将其解码为高斯参数；2）为进一步提升性能，提出混合高斯渲染，集成高效的体渲染设计用于新视角合成；3）支持特定场景快速微调，引入多视图几何一致性聚合策略，有效聚合可泛化模型生成的点云，作为逐场景优化的初始化。与以往基于NeRF的可泛化方法（通常需要数分钟微调和每张图像数秒渲染）相比，MVSGaussian实现了实时渲染，同时为每个场景提供更优的合成质量。与原始3D-GS相比，MVSGaussian以更低的训练计算成本实现了更好的视角合成。在DTU、Real Forward-facing、NeRF Synthetic及Tanks and Temples数据集上的大量实验验证了MVSGaussian在具有说服力的泛化性、实时渲染速度和快速逐场景优化方面达到了最先进性能。