SplineGS: Robust Motion-Adaptive Spline for Real-Time Dynamic 3D Gaussians from Monocular Video

TL;DR

SplineGS introduces a novel motion-adaptive spline framework for real-time, high-quality 3D scene reconstruction and rendering from monocular videos, effectively handling scene dynamics without multi-view cues.

Contribution

The paper presents SplineGS, a new monocular dynamic 3D Gaussian framework with a motion-adaptive spline and pruning strategy, eliminating the need for SfM preprocessing.

Findings

Outperforms state-of-the-art in novel view synthesis quality.

Achieves thousands of times faster rendering speed.

Robustly models dynamic scenes from monocular videos.

Abstract

Synthesizing novel views from in-the-wild monocular videos is challenging due to scene dynamics and the lack of multi-view cues. To address this, we propose SplineGS, a COLMAP-free dynamic 3D Gaussian Splatting (3DGS) framework for high-quality reconstruction and fast rendering from monocular videos. At its core is a novel Motion-Adaptive Spline (MAS) method, which represents continuous dynamic 3D Gaussian trajectories using cubic Hermite splines with a small number of control points. For MAS, we introduce a Motion-Adaptive Control points Pruning (MACP) method to model the deformation of each dynamic 3D Gaussian across varying motions, progressively pruning control points while maintaining dynamic modeling integrity. Additionally, we present a joint optimization strategy for camera parameter estimation and 3D Gaussian attributes, leveraging photometric and geometric consistency. This eliminates the need for Structure-from-Motion preprocessing and enhances SplineGS's robustness in real-world conditions. Experiments show that SplineGS significantly outperforms state-of-the-art methods in novel view synthesis quality for dynamic scenes from monocular videos, achieving thousands times faster rendering speed.