| title | Multi4D: High-Fidelity Dynamic Gaussian Splatting via Multi-Level Competitive Allocation |
|---|---|
| authors | Rui Wang, Quentin Lohmeyer, Siyu Tang, Mirko Meboldt |
| date | 2026-06-20 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2606.22197 |
| upvotes | 2 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | Dynamic 3D Gaussian splatting faces a fundamental tension between motion consistency and visual fidelity. Deformation-based approaches preserve temporal correspondence but suffer from motion over-factorization, oversmoothing high-frequency dynamics. In contrast, 4D-primitive methods capture fine visual details yet incur temporal overparameterization, breaking object identity and leading to severe storage overhead. To resolve this, we introduce Multi4D, a framework for high-fidelity dynamic Gaussian Splatting based on multi-level competitive allocation. Instead of a monolithic representation, we distribute modeling capacity across three structured levels: static structure, persistent dynamic geometry, and transient appearance primitives. Through shared rasterization and residual-driven optimization, these levels dynamically compete to explain photometric error, enabling adaptive specialization without pre-assigned decomposition. This allocation preserves long-term motion consistency while capturing fine dynamic detail, achieving state-of-the-art rendering quality and real-time performance with significantly fewer dynamic primitives. Furthermore, because our representation explicitly tracks compact persistent Gaussians over time, semantic features can be embedded afterward, enabling Multi4D to achieve state-of-the-art 4D segmentation accuracy with an order-of-magnitude speedup. Project page: https://batfacewayne.github.io/Multi4D.io/ |
cite
@misc{wang2026multi,
title={Multi4D: High-Fidelity Dynamic Gaussian Splatting via Multi-Level Competitive Allocation},
author={Rui Wang and Quentin Lohmeyer and Siyu Tang and Mirko Meboldt},
year={2026},
eprint={2606.22197},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2606.22197}
}