Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Hao, Li, Sicheng, Gao, Xiang, Batuer, Abudouaihati, Yu, Lu, Liao, Yiyi
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2505.07539
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908360330182656
author Li, Hao
Li, Sicheng
Gao, Xiang
Batuer, Abudouaihati
Yu, Lu
Liao, Yiyi
author_facet Li, Hao
Li, Sicheng
Gao, Xiang
Batuer, Abudouaihati
Yu, Lu
Liao, Yiyi
contents Immersive video offers a 6-Dof-free viewing experience, potentially playing a key role in future video technology. Recently, 4D Gaussian Splatting has gained attention as an effective approach for immersive video due to its high rendering efficiency and quality, though maintaining quality with manageable storage remains challenging. To address this, we introduce GIFStream, a novel 4D Gaussian representation using a canonical space and a deformation field enhanced with time-dependent feature streams. These feature streams enable complex motion modeling and allow efficient compression by leveraging temporal correspondence and motion-aware pruning. Additionally, we incorporate both temporal and spatial compression networks for end-to-end compression. Experimental results show that GIFStream delivers high-quality immersive video at 30 Mbps, with real-time rendering and fast decoding on an RTX 4090. Project page: https://xdimlab.github.io/GIFStream
format Preprint
id arxiv_https___arxiv_org_abs_2505_07539
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle GIFStream: 4D Gaussian-based Immersive Video with Feature Stream
Li, Hao
Li, Sicheng
Gao, Xiang
Batuer, Abudouaihati
Yu, Lu
Liao, Yiyi
Computer Vision and Pattern Recognition
Immersive video offers a 6-Dof-free viewing experience, potentially playing a key role in future video technology. Recently, 4D Gaussian Splatting has gained attention as an effective approach for immersive video due to its high rendering efficiency and quality, though maintaining quality with manageable storage remains challenging. To address this, we introduce GIFStream, a novel 4D Gaussian representation using a canonical space and a deformation field enhanced with time-dependent feature streams. These feature streams enable complex motion modeling and allow efficient compression by leveraging temporal correspondence and motion-aware pruning. Additionally, we incorporate both temporal and spatial compression networks for end-to-end compression. Experimental results show that GIFStream delivers high-quality immersive video at 30 Mbps, with real-time rendering and fast decoding on an RTX 4090. Project page: https://xdimlab.github.io/GIFStream
title GIFStream: 4D Gaussian-based Immersive Video with Feature Stream
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.07539