Temporally Aware Densification for Dynamic 3D Gaussian Splatting

2026-06-22

AI Summary

Existing dynamic 3D Gaussian splatting methods rely on static densification strategies, often resulting in blurry or incomplete reconstructions in dynamic regions. To address this limitation, this work proposes a temporally aware densification framework that adaptively optimizes Gaussian distributions for both static and dynamic scene content through three key mechanisms: Visibility-Aware Densification (VAD), Temporal Adaptive Thresholding (TAT), and Temporal Offset Warping (TOW). The proposed approach significantly outperforms current state-of-the-art methods across three dynamic multi-view benchmark datasets. Furthermore, it functions as a plug-and-play module that generalizes effectively to various dynamic 3D Gaussian splatting architectures, consistently enhancing the fidelity of dynamic scene reconstruction.

Abstract

Despite modeling temporal motion, dynamic 3D Gaussian Splatting (3DGS) methods still inherit a static densification strategy that is ill-suited for dynamic scenes. This neglect of temporal behavior leads to under-reconstructed and blurry dynamic regions, as short-lived Gaussians receive sparse supervision and fail to densify effectively. We propose a Visibility-Aware Densification (VAD) framework that integrates temporal visibility into the densification process, ensuring that Gaussians are refined based on their actual temporal presence. A Temporally-Adaptive Thresholding (TAT) mechanism further adjusts each Gaussian's densification threshold according to its temporal lifespan, promoting balanced refinement of both static and dynamic regions. Finally, a Temporal Offset Warping (TOW) design enhances deformation capacity around temporal centers, extending the lifespan of highly dynamic Gaussians and facilitating more effective densification. Our approach achieves substantial improvements in the visual quality of dynamic regions, outperforming existing methods across three dynamic multi-view benchmark datasets. Moreover, the proposed VAD module generalizes across diverse dynamic 3DGS methods, consistently improving dynamic reconstruction as a plug-and-play component.

Problem

• Dynamic 3D Gaussian Splatting
• temporal densification
• under-reconstruction
• blurry dynamic regions
• temporal visibility

Innovation

• Visibility-Aware Densification
• Temporally-Adaptive Thresholding
• Temporal Offset Warping
• Dynamic 3D Gaussian Splatting
• Temporal Densification