S2GO: Streaming Sparse Gaussian Occupancy Prediction

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Park, Jinhyung, Hu, Yihan, Peng, Chensheng, Zheng, Wenzhao, Kitani, Kris, Zhan, Wei
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916781312966656
author Park, Jinhyung
Hu, Yihan
Peng, Chensheng
Zheng, Wenzhao
Kitani, Kris
Zhan, Wei
author_facet Park, Jinhyung
Hu, Yihan
Peng, Chensheng
Zheng, Wenzhao
Kitani, Kris
Zhan, Wei
contents Despite the demonstrated efficiency and performance of sparse query-based representations for perception, state-of-the-art 3D occupancy prediction methods still rely on voxel-based or dense Gaussian-based 3D representations. However, dense representations are slow, and they lack flexibility in capturing the temporal dynamics of driving scenes. Distinct from prior work, we instead summarize the scene into a compact set of 3D queries which are propagated through time in an online, streaming fashion. These queries are then decoded into semantic Gaussians at each timestep. We couple our framework with a denoising rendering objective to guide the queries and their constituent Gaussians in effectively capturing scene geometry. Owing to its efficient, query-based representation, S2GO achieves state-of-the-art performance on the nuScenes and KITTI occupancy benchmarks, outperforming prior art (e.g., GaussianWorld) by 1.5 IoU with 5.9x faster inference.
format Preprint
id arxiv_https___arxiv_org_abs_2506_05473
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle S2GO: Streaming Sparse Gaussian Occupancy Prediction
Park, Jinhyung
Hu, Yihan
Peng, Chensheng
Zheng, Wenzhao
Kitani, Kris
Zhan, Wei
Computer Vision and Pattern Recognition
Despite the demonstrated efficiency and performance of sparse query-based representations for perception, state-of-the-art 3D occupancy prediction methods still rely on voxel-based or dense Gaussian-based 3D representations. However, dense representations are slow, and they lack flexibility in capturing the temporal dynamics of driving scenes. Distinct from prior work, we instead summarize the scene into a compact set of 3D queries which are propagated through time in an online, streaming fashion. These queries are then decoded into semantic Gaussians at each timestep. We couple our framework with a denoising rendering objective to guide the queries and their constituent Gaussians in effectively capturing scene geometry. Owing to its efficient, query-based representation, S2GO achieves state-of-the-art performance on the nuScenes and KITTI occupancy benchmarks, outperforming prior art (e.g., GaussianWorld) by 1.5 IoU with 5.9x faster inference.
title S2GO: Streaming Sparse Gaussian Occupancy Prediction
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.05473