FSGS: Real-Time Few-shot View Synthesis using Gaussian Splatting

By Zehao Zhu, Zhiwen Fan, Yifan Jiang, Zhangyang Wang. European Conference on Computer Vision, 2023

382 citations recorded by the source corpus.

Novel view synthesis from limited observations remains an important and persistent task. However, high efficiency in existing NeRF-based few-shot view synthesis is often compromised to obtain an accurate 3D representation. To address this challenge, we propose a few-shot view synthesis framework based on 3D Gaussian Splatting that enables real-time and photo-realistic view synthesis with as few as three training views. The proposed method, dubbed FSGS, handles the extremely sparse initialized SfM points with a thoughtfully designed Gaussian Unpooling process. Our method iteratively distributes new Gaussians around the most representative locations, subsequently infilling local details in vacant areas. We also integrate a large-scale pre-trained monocular depth estimator within the Gaussians optimization process, leveraging online augmented views to guide the geometric optimization towards an optimal solution. Starting from sparse points observed from limited input viewpoints, our FSGS can accurately grow into unseen regions, comprehensively covering the scene and boosting the rendering quality of novel views. Overall, FSGS achieves state-of-the-art performance in both accuracy and rendering efficiency across diverse datasets, including LLFF, Mip-NeRF360, and Blender. Project website: https://zehaozhu.github.io/FSGS/.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/51965f41228e35a798c55613c40dd1499584660b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Ying Shan
2. Jinbo Xing
3. Xin Ning
4. Xiaoyu Li
5. Wangbo Yu
6. Li Yuan
7. Xiangjun Gao
8. Wenguan Wang

Citations in the corpus

• Gen3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control (2025, neutral): “Several works proposed to address these issues through regularization using geometric priors [10, 38, 43, 48, 54, 63, 70, 75], which, however, are sensitive to noise in the estimated depth or normals.”
• ViewCrafter: Taming Video Diffusion Models for High-fidelity Novel View Synthesis (2024, methodology): “We compare our method with three 3D-GS representation-based sparse view reconstruction methods: DNGaussian [77], FSGS [78] and InstantSplat [79].”
• DNGaussian: Optimizing Sparse-View 3D Gaussian Radiance Fields with Global-Local Depth Normalization (2024, background): “To tackle the scale ambiguity and error of monocular depths, previous works have introduced various scale-invariant losses [10, 36, 55] including depth ranking loss [42, 50], however, all of which are not optimal for us.”
• A Survey on 3D Gaussian Splatting (2024, positive): “To provide a synthesized overview before delving into each direction, Table S1 summarizes their key ideas, representative works (ordered chronologically), advantages, and limitations.”
• Recent advances in 3D Gaussian splatting (2024, neutral): “The most straightforward way is to assign different attribute values to 3D Gaussian at different timesteps.”