SparseCoop: Cooperative Perception with Kinematic-Grounded Queries

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Jiahao, Jiang, Zhongwei, Sun, Wenchao, Zhong, Jiaru, Yu, Haibao, Zhang, Yuner, Lu, Chenyang, Zhang, Chuang, He, Lei, Xu, Shaobing, Wang, Jianqiang
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910117958516736
author Wang, Jiahao
Jiang, Zhongwei
Sun, Wenchao
Zhong, Jiaru
Yu, Haibao
Zhang, Yuner
Lu, Chenyang
Zhang, Chuang
He, Lei
Xu, Shaobing
Wang, Jianqiang
author_facet Wang, Jiahao
Jiang, Zhongwei
Sun, Wenchao
Zhong, Jiaru
Yu, Haibao
Zhang, Yuner
Lu, Chenyang
Zhang, Chuang
He, Lei
Xu, Shaobing
Wang, Jianqiang
contents Cooperative perception is critical for autonomous driving, overcoming the inherent limitations of a single vehicle, such as occlusions and constrained fields-of-view. However, current approaches sharing dense Bird's-Eye-View (BEV) features are constrained by quadratically-scaling communication costs and the lack of flexibility and interpretability for precise alignment across asynchronous or disparate viewpoints. While emerging sparse query-based methods offer an alternative, they often suffer from inadequate geometric representations, suboptimal fusion strategies, and training instability. In this paper, we propose SparseCoop, a fully sparse cooperative perception framework for 3D detection and tracking that completely discards intermediate BEV representations. Our framework features a trio of innovations: a kinematic-grounded instance query that uses an explicit state vector with 3D geometry and velocity for precise spatio-temporal alignment; a coarse-to-fine aggregation module for robust fusion; and a cooperative instance denoising task to accelerate and stabilize training. Experiments on V2X-Seq and Griffin datasets show SparseCoop achieves state-of-the-art performance. Notably, it delivers this with superior computational efficiency, low transmission cost, and strong robustness to communication latency. Code is available at https://github.com/wang-jh18-SVM/SparseCoop.
format Preprint
id arxiv_https___arxiv_org_abs_2512_06838
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SparseCoop: Cooperative Perception with Kinematic-Grounded Queries
Wang, Jiahao
Jiang, Zhongwei
Sun, Wenchao
Zhong, Jiaru
Yu, Haibao
Zhang, Yuner
Lu, Chenyang
Zhang, Chuang
He, Lei
Xu, Shaobing
Wang, Jianqiang
Computer Vision and Pattern Recognition
Cooperative perception is critical for autonomous driving, overcoming the inherent limitations of a single vehicle, such as occlusions and constrained fields-of-view. However, current approaches sharing dense Bird's-Eye-View (BEV) features are constrained by quadratically-scaling communication costs and the lack of flexibility and interpretability for precise alignment across asynchronous or disparate viewpoints. While emerging sparse query-based methods offer an alternative, they often suffer from inadequate geometric representations, suboptimal fusion strategies, and training instability. In this paper, we propose SparseCoop, a fully sparse cooperative perception framework for 3D detection and tracking that completely discards intermediate BEV representations. Our framework features a trio of innovations: a kinematic-grounded instance query that uses an explicit state vector with 3D geometry and velocity for precise spatio-temporal alignment; a coarse-to-fine aggregation module for robust fusion; and a cooperative instance denoising task to accelerate and stabilize training. Experiments on V2X-Seq and Griffin datasets show SparseCoop achieves state-of-the-art performance. Notably, it delivers this with superior computational efficiency, low transmission cost, and strong robustness to communication latency. Code is available at https://github.com/wang-jh18-SVM/SparseCoop.
title SparseCoop: Cooperative Perception with Kinematic-Grounded Queries
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.06838