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

WACV 2026
1National Yang Ming Chiao Tung University, Taiwan
2National Chung Cheng University, Taiwan
MY ALT TEXT

Our TED-4DGS reconstructs the scene with superior rendering quality compared to ADC-GS. It achieves a 26% file size reduction while closely matching the ground-truth view.

Right: Temporal duration map. Static background regions reuse long-duration Gaussian primitives, whereas occluded parts of the hand and banana are represented by short-duration primitives, demonstrating the effectiveness of ours temporal activation method.

Abstract

Building on the success of 3D Gaussian Splatting (3DGS), representing dynamic scenes (4DGS) has become a key research focus. However, achieving efficient deformation modeling while simultaneously optimizing for compression (Rate-Distortion Optimization) remains a significant challenge. Prior methods often struggle with either overspecified, short-lived primitives or irregular deformations that lack explicit temporal control.

To address this, we present TED-4DGS, a novel framework that unifies Temporally Activated mechanisms with Embedding-based Deformation for high-efficiency 4DGS compression. Our approach introduces learnable temporal activation parameters for each anchor to explicitly model appearance and disappearance, effectively handling occlusion and disocclusion. Furthermore, we incorporate an INR-based hyperprior and a channel-wise autoregressive model to achieve compact attribute compression. Experimental results demonstrate that TED-4DGS achieves state-of-the-art rate-distortion performance on real-world datasets, significantly reducing file sizes while maintaining superior rendering quality compared to existing methods.

Main Method

Our work, TED-4DGS, introduces a rate-distortion-optimized compression framework for dynamic 3DGS. We design a per-anchor embedding-based deformation network (a) that leverages temporal features to query a shared global deformation bank to effectively capture anchor-specific deformation. To promote stable deformation and improved visibility modeling, we extend static 3D anchors into 4D by introducing temporal activation parameters. Finally, we incorporate an INR-based hyperprior and a channel-wise autoregressive model (b) for efficient attribute coding.

Overview of CAT-3DGS framework

Overall pipeline of our TED-4DGS framework.

Overview of TED-4 DGS framework

(a) Anchor-based deformation framework and Scaffold-GS rendering.

Overview of TED-4 DGS framework

(b) INR-based anchor attribute compression framework.

Results

Our TED-4DGS consistently outperforms the competing methods across both datasets(HyperNeRF and Neu3D), achieving the state-of-the-art RD performance.

MY ALT TEXT MY ALT TEXT

Visual Comparison

HyperNeRF Dataset

Ours
Ground Truth
Banana scene(28.3dB/ 5.1MB)
Ours
Ground Truth
Chicken scene (29.3dB/ 2.09MB)

Neu3D Dataset

Ground Truth
Ours
Cut Roasted Beef (33.4dB/ 2.0MB)
Ground Truth
Ours
Flame Steak (33.0dB/ 1.8MB)

Temporal Duration Video on HyperNeRF Banana Scene

Time Duration

Short Long

BibTeX

@misc{ho2025ted4dgstemporallyactivatedembeddingbased,
      title={TED-4DGS: Temporally Activated and Embedding-based Deformation for 4DGS Compression}, 
      author={Cheng-Yuan Ho and He-Bi Yang and Jui-Chiu Chiang and Yu-Lun Liu and Wen-Hsiao Peng},
      year={2025},
      eprint={2512.05446},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2512.05446}, 
}
  

Acknowledgement

This work is supported by MediaTek Advanced Research Center and National Science and Technology Council (NSTC), Taiwan, under Grants 113-2634-F-A49-007-, 112-2221-E-A49-092-MY3, and 114-2221-E-A49-035-MY3. We thank National Center for High-performance Computing (NCHC) for providing computational and storage resources.