ObjectGS: Object-aware Scene Reconstruction and Scene Understanding via Gaussian Splatting

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhu, Ruijie, Yu, Mulin, Xu, Linning, Jiang, Lihan, Li, Yixuan, Zhang, Tianzhu, Pang, Jiangmiao, Dai, Bo
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915402542481408
author Zhu, Ruijie
Yu, Mulin
Xu, Linning
Jiang, Lihan
Li, Yixuan
Zhang, Tianzhu
Pang, Jiangmiao
Dai, Bo
author_facet Zhu, Ruijie
Yu, Mulin
Xu, Linning
Jiang, Lihan
Li, Yixuan
Zhang, Tianzhu
Pang, Jiangmiao
Dai, Bo
contents 3D Gaussian Splatting is renowned for its high-fidelity reconstructions and real-time novel view synthesis, yet its lack of semantic understanding limits object-level perception. In this work, we propose ObjectGS, an object-aware framework that unifies 3D scene reconstruction with semantic understanding. Instead of treating the scene as a unified whole, ObjectGS models individual objects as local anchors that generate neural Gaussians and share object IDs, enabling precise object-level reconstruction. During training, we dynamically grow or prune these anchors and optimize their features, while a one-hot ID encoding with a classification loss enforces clear semantic constraints. We show through extensive experiments that ObjectGS not only outperforms state-of-the-art methods on open-vocabulary and panoptic segmentation tasks, but also integrates seamlessly with applications like mesh extraction and scene editing. Project page: https://ruijiezhu94.github.io/ObjectGS_page
format Preprint
id arxiv_https___arxiv_org_abs_2507_15454
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ObjectGS: Object-aware Scene Reconstruction and Scene Understanding via Gaussian Splatting
Zhu, Ruijie
Yu, Mulin
Xu, Linning
Jiang, Lihan
Li, Yixuan
Zhang, Tianzhu
Pang, Jiangmiao
Dai, Bo
Graphics
Artificial Intelligence
Computer Vision and Pattern Recognition
Human-Computer Interaction
3D Gaussian Splatting is renowned for its high-fidelity reconstructions and real-time novel view synthesis, yet its lack of semantic understanding limits object-level perception. In this work, we propose ObjectGS, an object-aware framework that unifies 3D scene reconstruction with semantic understanding. Instead of treating the scene as a unified whole, ObjectGS models individual objects as local anchors that generate neural Gaussians and share object IDs, enabling precise object-level reconstruction. During training, we dynamically grow or prune these anchors and optimize their features, while a one-hot ID encoding with a classification loss enforces clear semantic constraints. We show through extensive experiments that ObjectGS not only outperforms state-of-the-art methods on open-vocabulary and panoptic segmentation tasks, but also integrates seamlessly with applications like mesh extraction and scene editing. Project page: https://ruijiezhu94.github.io/ObjectGS_page
title ObjectGS: Object-aware Scene Reconstruction and Scene Understanding via Gaussian Splatting
topic Graphics
Artificial Intelligence
Computer Vision and Pattern Recognition
Human-Computer Interaction
url https://arxiv.org/abs/2507.15454