ACA-GS: Adaptive-Capacity Anchored Gaussian Splatting for Compact Dynamic Radiance Fields

2026-08-05

AI Summary

This work addresses the inherent trade-off between motion expressiveness and storage efficiency in 4D Gaussian splatting by introducing an adaptive anchor framework that dynamically allocates geometric and appearance representation resources according to local spatiotemporal demands. The proposed method uniquely combines adaptive anchor cardinality with feature channel masking, enabling on-the-fly adjustment of the number of Gaussians and their active feature dimensions within localized regions. This approach achieves significantly improved compression efficiency without compromising rendering quality. Experimental results on standard benchmarks such as MPEG demonstrate that the method attains up to 1.5× higher compression ratios compared to state-of-the-art techniques while maintaining comparable visual fidelity.

Abstract

Recent advances in 4D Gaussian Splatting (4DGS) enable high-fidelity, real-time spatiotemporal rendering, but expose a fundamental trade-off between motion expressiveness and storage efficiency. While anchor-based designs achieve compactness through anchor-level parameter sharing, their rigid uniform parametrization enforces fixed Neural Gaussian counts and feature budgets per anchor. Consequently, insufficient fidelity is addressed by excessive anchor density, rather than lightweight, targeted increases in Neural Gaussian count or feature capacity, resulting in memory waste. To overcome this rigidity, we introduce an adaptive-capacity anchor-based framework that dynamically allocates the representational capacity based on local spatiotemporal demands. Adaptive Anchor Cardinality varies the number of Neural Gaussians per anchor, concentrating primitives in regions of high geometric or motion complexity while suppressing redundancy. In parallel, Adaptive Anchor Feature Masking modulates anchor-level feature channels, assigning rich features to complex regions and lightweight representations to simpler ones. Experiments on MPEG, Panoptic Sports, and N3DV datasets demonstrate substantial storage reduction without degrading visual quality. Notably, on challenging MPEG sequences with complex motion, our method achieves up to 1.5x higher compression than state-of-the-art anchor-based methods while preserving comparable quality.

Problem

• 4D Gaussian Splatting
• dynamic radiance fields
• storage efficiency
• motion expressiveness
• anchor-based representation

Innovation

• Adaptive-Capacity
• Gaussian Splatting
• Dynamic Radiance Fields
• Anchor-Based Representation
• Spatiotemporal Compression