3D Gaussian Representations with Motion Trajectory Field for Dynamic Scene Reconstruction

2025-08-10

AI Summary

To address the challenges of novel-view synthesis and motion reconstruction in monocular dynamic scenes, this paper proposes a decoupled Dynamic 3D Gaussian Splatting (Dynamic 3DGS) framework. Our core innovation is a motion trajectory field that jointly represents object motion via time-invariant motion coefficients and shared trajectory bases, explicitly decoupling dynamic foreground from static background. This design preserves the rendering efficiency of 3DGS while significantly improving the physical plausibility and optimization stability of motion modeling. We adopt a staged, decoupled training strategy to jointly optimize geometry, appearance, and motion parameters. Extensive experiments on complex dynamic scenes demonstrate state-of-the-art performance in both novel-view synthesis quality (PSNR/SSIM) and motion trajectory reconstruction accuracy. Moreover, our method enhances spatiotemporal consistency and physical fidelity of the reconstructed 3D scene.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: Motion and Path Planning
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

This paper addresses the challenge of novel-view synthesis and motion reconstruction of dynamic scenes from monocular video, which is critical for many robotic applications. Although Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have demonstrated remarkable success in rendering static scenes, extending them to reconstruct dynamic scenes remains challenging. In this work, we introduce a novel approach that combines 3DGS with a motion trajectory field, enabling precise handling of complex object motions and achieving physically plausible motion trajectories. By decoupling dynamic objects from static background, our method compactly optimizes the motion trajectory field. The approach incorporates time-invariant motion coefficients and shared motion trajectory bases to capture intricate motion patterns while minimizing optimization complexity. Extensive experiments demonstrate that our approach achieves state-of-the-art results in both novel-view synthesis and motion trajectory recovery from monocular video, advancing the capabilities of dynamic scene reconstruction.

Problem

• Novel-view synthesis for dynamic scenes from monocular video
• Reconstructing physically plausible motion trajectories in dynamic scenes
• Decoupling dynamic objects from static background for optimized reconstruction

Innovation

• Combines 3DGS with motion trajectory field
• Decouples dynamic objects from static background
• Uses time-invariant motion coefficients