MGStream: Motion-aware 3D Gaussian for Streamable Dynamic Scene Reconstruction

TL;DR

MGStream introduces motion-aware 3D Gaussians to improve dynamic scene reconstruction, reducing flickering artifacts and enhancing storage efficiency in streamable 3D scene synthesis.

Contribution

It proposes a novel motion-related 3D Gaussian approach that models dynamic and static scene components separately, enabling better handling of emerging objects and dynamic changes.

Findings

Outperforms existing methods in rendering quality.

Reduces flickering artifacts and improves temporal consistency.

Enhances storage and training efficiency.

Abstract

3D Gaussian Splatting (3DGS) has gained significant attention in streamable dynamic novel view synthesis (DNVS) for its photorealistic rendering capability and computational efficiency. Despite much progress in improving rendering quality and optimization strategies, 3DGS-based streamable dynamic scene reconstruction still suffers from flickering artifacts and storage inefficiency, and struggles to model the emerging objects. To tackle this, we introduce MGStream which employs the motion-related 3D Gaussians (3DGs) to reconstruct the dynamic and the vanilla 3DGs for the static. The motion-related 3DGs are implemented according to the motion mask and the clustering-based convex hull algorithm. The rigid deformation is applied to the motion-related 3DGs for modeling the dynamic, and the attention-based optimization on the motion-related 3DGs enables the reconstruction of the emerging objects. As the deformation and optimization are only conducted on the motion-related 3DGs, MGStream avoids flickering artifacts and improves the storage efficiency. Extensive experiments on real-world datasets N3DV and MeetRoom demonstrate that MGStream surpasses existing streaming 3DGS-based approaches in terms of rendering quality, training/storage efficiency and temporal consistency. Our code is available at: https://github.com/pcl3dv/MGStream.