Flow Distillation Sampling: Regularizing 3D Gaussians with Pre-trained Matching Priors

TL;DR

This paper introduces Flow Distillation Sampling (FDS), a novel technique that uses pre-trained geometric priors and optical flow to improve 3D Gaussian Splatting's geometric reconstruction and rendering quality, especially in sparse view regions.

Contribution

The paper proposes FDS, a new sampling method that incorporates pre-trained matching priors and optical flow to enhance 3D Gaussian Splatting's geometric accuracy and rendering performance.

Findings

FDS improves geometric reconstruction in sparse view regions.

FDS enhances rendering quality and view synthesis accuracy.

Experimental results outperform state-of-the-art methods.

Abstract

3D Gaussian Splatting (3DGS) has achieved excellent rendering quality with fast training and rendering speed. However, its optimization process lacks explicit geometric constraints, leading to suboptimal geometric reconstruction in regions with sparse or no observational input views. In this work, we try to mitigate the issue by incorporating a pre-trained matching prior to the 3DGS optimization process. We introduce Flow Distillation Sampling (FDS), a technique that leverages pre-trained geometric knowledge to bolster the accuracy of the Gaussian radiance field. Our method employs a strategic sampling technique to target unobserved views adjacent to the input views, utilizing the optical flow calculated from the matching model (Prior Flow) to guide the flow analytically calculated from the 3DGS geometry (Radiance Flow). Comprehensive experiments in depth rendering, mesh reconstruction, and novel view synthesis showcase the significant advantages of FDS over state-of-the-art methods. Additionally, our interpretive experiments and analysis aim to shed light on the effects of FDS on geometric accuracy and rendering quality, potentially providing readers with insights into its performance. Project page: https://nju-3dv.github.io/projects/fds