SQS: Enhancing Sparse Perception Models via Query-based Splatting in Autonomous Driving

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhang, Haiming, Zhu, Yiyao, Zhou, Wending, Yan, Xu, Cai, Yingjie, Liu, Bingbing, Cui, Shuguang, Li, Zhen
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915504648617984
author Zhang, Haiming
Zhu, Yiyao
Zhou, Wending
Yan, Xu
Cai, Yingjie
Liu, Bingbing
Cui, Shuguang
Li, Zhen
author_facet Zhang, Haiming
Zhu, Yiyao
Zhou, Wending
Yan, Xu
Cai, Yingjie
Liu, Bingbing
Cui, Shuguang
Li, Zhen
contents Sparse Perception Models (SPMs) adopt a query-driven paradigm that forgoes explicit dense BEV or volumetric construction, enabling highly efficient computation and accelerated inference. In this paper, we introduce SQS, a novel query-based splatting pre-training specifically designed to advance SPMs in autonomous driving. SQS introduces a plug-in module that predicts 3D Gaussian representations from sparse queries during pre-training, leveraging self-supervised splatting to learn fine-grained contextual features through the reconstruction of multi-view images and depth maps. During fine-tuning, the pre-trained Gaussian queries are seamlessly integrated into downstream networks via query interaction mechanisms that explicitly connect pre-trained queries with task-specific queries, effectively accommodating the diverse requirements of occupancy prediction and 3D object detection. Extensive experiments on autonomous driving benchmarks demonstrate that SQS delivers considerable performance gains across multiple query-based 3D perception tasks, notably in occupancy prediction and 3D object detection, outperforming prior state-of-the-art pre-training approaches by a significant margin (i.e., +1.3 mIoU on occupancy prediction and +1.0 NDS on 3D detection).
format Preprint
id arxiv_https___arxiv_org_abs_2509_16588
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SQS: Enhancing Sparse Perception Models via Query-based Splatting in Autonomous Driving
Zhang, Haiming
Zhu, Yiyao
Zhou, Wending
Yan, Xu
Cai, Yingjie
Liu, Bingbing
Cui, Shuguang
Li, Zhen
Computer Vision and Pattern Recognition
Artificial Intelligence
Robotics
Sparse Perception Models (SPMs) adopt a query-driven paradigm that forgoes explicit dense BEV or volumetric construction, enabling highly efficient computation and accelerated inference. In this paper, we introduce SQS, a novel query-based splatting pre-training specifically designed to advance SPMs in autonomous driving. SQS introduces a plug-in module that predicts 3D Gaussian representations from sparse queries during pre-training, leveraging self-supervised splatting to learn fine-grained contextual features through the reconstruction of multi-view images and depth maps. During fine-tuning, the pre-trained Gaussian queries are seamlessly integrated into downstream networks via query interaction mechanisms that explicitly connect pre-trained queries with task-specific queries, effectively accommodating the diverse requirements of occupancy prediction and 3D object detection. Extensive experiments on autonomous driving benchmarks demonstrate that SQS delivers considerable performance gains across multiple query-based 3D perception tasks, notably in occupancy prediction and 3D object detection, outperforming prior state-of-the-art pre-training approaches by a significant margin (i.e., +1.3 mIoU on occupancy prediction and +1.0 NDS on 3D detection).
title SQS: Enhancing Sparse Perception Models via Query-based Splatting in Autonomous Driving
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Robotics
url https://arxiv.org/abs/2509.16588