MVSGaussian: fast generalizable Gaussian splatting reconstruction from multi-view stereo

• Main Authors: Liu, Tianqi, Wang, Guangcong, Hu, Shoukang, Shen, Liao, Ye, Xinyi, Zang, Yuhang, Cao, Zhiguo, Li, Wei, Liu, Ziwei
• Other Authors: College of Computing and Data Science
• Format: Conference or Workshop Item
• Language: English
• Published: 2025
• Subjects: Computer and Information Science; Generalizable Gaussian splatting; Multi-view stereo
• Online Access: https://hdl.handle.net/10356/182591; http://arxiv.org/abs/2405.12218v3
• Institution: Nanyang Technological University

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.