arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.08688cs.CVcs.ROeess.IV
收藏

GenMapping:释放逆透视变换的潜力以实现鲁棒的在线高清地图构建

GenMapping: Unleashing the Potential of Inverse Perspective Mapping for Robust Online HD Map Construction

National Engineering Research Center of Robot Visual Perception and Control Technology(机器人视觉感知与控制技术国家工程研究中心);State Key Laboratory of Extreme Photonics and Instrumentation, Zhejiang University(浙江大学极端光学仪器与技术国家重点实验室);Shanghai Supremind Technology Company Ltd(上海赛曼德科技有限公司);College of Computer Science, Zhejiang University(浙江大学计算机学院);USC Viterbi School of Engineering, the University of Southern California(南加州大学维特比工程学院)

Siyu Li, Kailun Yang, Hao Shi, Song Wang, You Yao, Zhiyong Li

AI总结:

针对现有在线高清地图模型泛化性差的问题,提出基于逆透视变换的GenMapping三元协同框架,结合多分支特征与跨视图学习、双向增强,在两类建图任务上超越SOTA且推理速度快。

摘要:

在线高清(HD)地图凭借灵活的更新能力和更低的维护成本,已成为自动驾驶领域的首选方案,其表现盖过了离线高清地图这一同类方案。然而,当前的在线高清地图模型将视觉传感器参数嵌入训练过程,导致在应用于参数不同的视觉传感器时,泛化性能大幅下降。受逆透视变换(Inverse Perspective Mapping, IPM)将相机参数与训练过程解耦的固有潜力启发,我们设计了一个通用的地图生成框架GenMapping。该框架采用三元协同架构构建,包含主分支和两个辅助分支。当面对经IPM转换后存在局部畸变的粗糙道路图像时,主分支在状态空间模型下学习鲁棒的全局特征。两个辅助分支分别是稠密透视分支和稀疏先验分支:前者挖掘静态与动态物体之间的关联信息,后者引入OpenStreetMap(OSM,开放街道地图)的先验知识。我们设计了三重增强融合模块,以协同整合三个分支的独特空间特征。为进一步提升泛化能力,研究采用跨视图地图学习(Cross-View Map Learning, CVML)方案,在公共空间内实现联合学习。此外,还引入了双向数据增强(Bidirectional Data Augmentation, BiDA)模块,同时降低对数据集的依赖。大量实验结果表明,所提出的模型在语义建图和矢量化建图任务上均超越了当前的最优方法,同时还保持了较快的推理速度。源代码将公开发布于https://github.com/lynn-yu/GenMapping。

英文摘要:

Online High-Definition (HD) maps have emerged as the preferred option for autonomous driving, overshadowing the counterpart offline HD maps due to flexible update capability and lower maintenance costs. However, contemporary online HD map models embed parameters of visual sensors into training, resulting in a significant decrease in generalization performance when applied to visual sensors with different parameters. Inspired by the inherent potential of Inverse Perspective Mapping (IPM), where camera parameters are decoupled from the training process, we have designed a universal map generation framework, GenMapping. The framework is established with a triadic synergy architecture, including principal and dual auxiliary branches. When faced with a coarse road image with local distortion translated via IPM, the principal branch learns robust global features under the state space models. The two auxiliary branches are a dense perspective branch and a sparse prior branch. The former exploits the correlation information between static and moving objects, whereas the latter introduces the prior knowledge of OpenStreetMap (OSM). The triple-enhanced merging module is crafted to synergistically integrate the unique spatial features from all three branches. To further improve generalization capabilities, a Cross-View Map Learning (CVML) scheme is leveraged to realize joint learning within the common space. Additionally, a Bidirectional Data Augmentation (BiDA) module is introduced to mitigate reliance on datasets concurrently. A thorough array of experimental results shows that the proposed model surpasses current state-of-the-art methods in both semantic mapping and vectorized mapping, while also maintaining a rapid inference speed. The source code will be publicly available at https://github.com/lynn-yu/GenMapping.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑