TED-4DGS: Temporally Activated and Embedding-based Deformation for 4DGS Compression

TL;DR

TED-4DGS introduces a novel temporally activated and embedding-based deformation scheme for 4D Gaussian Splatting, enabling efficient, rate-distortion-optimized compression of dynamic 3D scene representations with state-of-the-art results.

Contribution

It proposes a unified deformation and compression framework for 4DGS that combines temporal activation, embedding-based deformation, and neural priors for improved efficiency.

Findings

Achieves state-of-the-art rate-distortion performance on real-world datasets.

Unifies strengths of existing 4DGS methods with novel temporal activation and embedding.

First to optimize compression for dynamic 3D Gaussian Splatting representations.

Abstract

Building on the success of 3D Gaussian Splatting (3DGS) in static 3D scene representation, its extension to dynamic scenes, commonly referred to as 4DGS or dynamic 3DGS, has attracted increasing attention. However, designing more compact and efficient deformation schemes together with rate-distortion-optimized compression strategies for dynamic 3DGS representations remains an underexplored area. Prior methods either rely on space-time 4DGS with overspecified, short-lived Gaussian primitives or on canonical 3DGS with deformation that lacks explicit temporal control. To address this, we present TED-4DGS, a temporally activated and embedding-based deformation scheme for rate-distortion-optimized 4DGS compression that unifies the strengths of both families. TED-4DGS is built on a sparse anchor-based 3DGS representation. Each canonical anchor is assigned learnable temporal-activation parameters to specify its appearance and disappearance transitions over time, while a lightweight per-anchor temporal embedding queries a shared deformation bank to produce anchor-specific deformation. For rate-distortion compression, we incorporate an implicit neural representation (INR)-based hyperprior to model anchor attribute distributions, along with a channel-wise autoregressive model to capture intra-anchor correlations. With these novel elements, our scheme achieves state-of-the-art rate-distortion performance on several real-world datasets. To the best of our knowledge, this work represents one of the first attempts to pursue a rate-distortion-optimized compression framework for dynamic 3DGS representations.