Generalizable 3D Gaussian Splatting enabled Semantic Coding for Real-Time Immersive Video Communications

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Dingxi, Guo, Wenqi, Liu, Yue, Han, Jungong, Qin, Zhijin
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915962747355136
author Yang, Dingxi
Guo, Wenqi
Liu, Yue
Han, Jungong
Qin, Zhijin
author_facet Yang, Dingxi
Guo, Wenqi
Liu, Yue
Han, Jungong
Qin, Zhijin
contents Real-time immersive video communications, particularly high-fidelity 3D telepresence, necessitates a synergistic balance between instantaneous dynamic scene reconstruction and high-efficiency data transmission. While recent advancements in feed-forward 3D Gaussian Splatting (3DGS) have enabled real-time rendering, performing multi-view video coding and 3D reconstruction in a decoupled manner leads to suboptimal compression efficiency and high computational complexity. To address this, we propose GS-SCNet, the first unified end-to-end framework that seamlessly integrates generalizable 3DGS reconstruction with a dedicated deep Semantic Coding pipeline. Our architecture is underpinned by two core technical contributions: (i) we introduce a Disparity-Guided Parallel Semantic Codec that exploits epipolar geometric priors to facilitate cross-view contextual interaction via disparity compensation and semantic fusion, thereby enabling real-time parallel processing of stereo streams while significantly enhancing rate-distortion performance, and (ii) we develop a Lightweight Gaussian Parameter Predictor which directly projects decoded semantic latents into 3DGS attributes, obviating the need for intermediate pixel-domain reconstruction. By coupling the codec with the task-specific predictor, our framework extracts geometric correlations only once, effectively eliminating the redundant computational bottleneck inherent in conventional decoupled paradigms. Extensive evaluations on both synthetic and real-world human datasets demonstrate that GS-SCNet achieves a superior trade-off across compression efficiency, rendering quality, and real-time performance. Notably, our framework exhibits strong cross-domain generalization and robustness against compression artifacts when applied to out-of-domain real-world data, significantly outperforming conventional decoupled transmission paradigms.
format Preprint
id arxiv_https___arxiv_org_abs_2604_25330
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Generalizable 3D Gaussian Splatting enabled Semantic Coding for Real-Time Immersive Video Communications
Yang, Dingxi
Guo, Wenqi
Liu, Yue
Han, Jungong
Qin, Zhijin
Image and Video Processing
Real-time immersive video communications, particularly high-fidelity 3D telepresence, necessitates a synergistic balance between instantaneous dynamic scene reconstruction and high-efficiency data transmission. While recent advancements in feed-forward 3D Gaussian Splatting (3DGS) have enabled real-time rendering, performing multi-view video coding and 3D reconstruction in a decoupled manner leads to suboptimal compression efficiency and high computational complexity. To address this, we propose GS-SCNet, the first unified end-to-end framework that seamlessly integrates generalizable 3DGS reconstruction with a dedicated deep Semantic Coding pipeline. Our architecture is underpinned by two core technical contributions: (i) we introduce a Disparity-Guided Parallel Semantic Codec that exploits epipolar geometric priors to facilitate cross-view contextual interaction via disparity compensation and semantic fusion, thereby enabling real-time parallel processing of stereo streams while significantly enhancing rate-distortion performance, and (ii) we develop a Lightweight Gaussian Parameter Predictor which directly projects decoded semantic latents into 3DGS attributes, obviating the need for intermediate pixel-domain reconstruction. By coupling the codec with the task-specific predictor, our framework extracts geometric correlations only once, effectively eliminating the redundant computational bottleneck inherent in conventional decoupled paradigms. Extensive evaluations on both synthetic and real-world human datasets demonstrate that GS-SCNet achieves a superior trade-off across compression efficiency, rendering quality, and real-time performance. Notably, our framework exhibits strong cross-domain generalization and robustness against compression artifacts when applied to out-of-domain real-world data, significantly outperforming conventional decoupled transmission paradigms.
title Generalizable 3D Gaussian Splatting enabled Semantic Coding for Real-Time Immersive Video Communications
topic Image and Video Processing
url https://arxiv.org/abs/2604.25330