Search papers, labs, and topics across Lattice.
This paper introduces WebSpline, a novel framework that leverages a Structure-Informed Spline (SIS) representation to achieve high-fidelity dynamic 3D Gaussian reconstruction from monocular videos. By utilizing a Structural Proxy Graph (SPG) to maintain structural coherence and applying a two-stage optimization process, WebSpline effectively balances global structure and local detail while enabling rapid rendering. Experimental results show that WebSpline outperforms existing methods, achieving state-of-the-art rendering quality and rendering speed over 10 times faster than the second-best method on the iPhone dataset.
WebSpline achieves unprecedented rendering speed and quality in dynamic 3D scene reconstruction from monocular videos, outperforming existing techniques by a significant margin.
Dynamic scene reconstruction from monocular videos remains highly challenging, as existing methods often struggle to balance global structural coherence and local fine-grained details under limited multi-view cues. To address this challenge, we propose WebSpline, a novel dynamic 3D Gaussian framework that enables structurally coherent and high-fidelity reconstruction from monocular videos with fast rendering. The core of WebSpline is the Structure-Informed Spline (SIS) representation, which models each dynamic Gaussian trajectory using a learnable cubic Hermite spline whose motion is structurally organized with an auxiliary Structural Proxy Graph (SPG). The proposed framework is optimized in two stages: (i) in the first stage, the SPG is initialized from 2D point tracks and refined with temporal rigidity regularization to establish structural coherence for moving objects across the sequence; and (ii) in the second stage, the SIS representation is initialized from the refined SPG and optimized under both spatial and structural neighborhood constraints. At inference, Gaussian motion is obtained solely by evaluating the learned SIS, enabling fast rendering. Extensive experiments on the challenging monocular dynamic scene benchmarks, iPhone and NVIDIA, demonstrate that our WebSpline achieves state-of-the-art rendering quality while rendering over 10 times faster than WorldTree, the second-best method on the iPhone dataset.
