虚拟点增强与双向跨注意力融合的3D目标检测
CSTR:
作者:
作者单位:

(1.华北计算机系统工程研究所,北京 100000; 2.山东大学 网络空间安全学院(研究院),山东 青岛 266237; 3.北京航空航天大学 网络空间安全学院,北京 100191)

作者简介:

马保全(1982—),男,博士, 正高级工程师

通讯作者:

中图分类号:

TP391.41

基金项目:

国家自然科学基金(U2241213)


Virtual point-enhanced 3D object detection with bidirectional cross-attention fusion
Author:
Affiliation:

(1.North China Computer System Engineering Research Institute, Beijing 100000, China; 2.School of Cyber Science and Technology, Shandong University, Qingdao 266237, Shandong, China; 3.School of Cyber Science and Technology, Beihang University, Beijing 100191, China)

Fund Project:

undefined

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为解决激光雷达(LiDAR)点云稀疏导致远距离目标几何信息缺失,以及现有模态融合方法中存在的投影噪声干扰和特征交互非标准化问题,提出了一种端到端的多模态三维检测网络——VAF-Net。首先,构建了抗噪的虚拟点增强框架,利用图像生成虚拟点以稠密化原始点云,并采用VirConvNet作为骨干网络,在提取全局几何特征的同时有效过滤由虚拟点引入的噪声。其次,设计了色彩语义特征提取支路,通过语义提升模块精准提取与三维空间对齐的图像局部纹理与上下文语义。最后,提出了基于标准Transformer原则的双向跨注意力融合模块,通过对称的双流架构和门控机制,实现了LiDAR几何特征与图像语义特征的深度、稳定且双向的交互。在KITTI三维目标检测基准上的实验表明,VAF-Net取得了当前最优性能。特别是在困难样本检测中,其3D平均精度达到82.74%,显著优于Voxel-RCNN等主流方法。研究结果表明,VAF-Net通过标准化的双向融合架构,有效实现了多模态信息的互补增强,大幅提升了复杂场景下的检测鲁棒性。

    Abstract:

    To address the loss of geometric information of distant objects caused by the sparsity of light detection and ranging (LiDAR) point clouds and the problems of projection noise interference and non-standardized feature interaction in existing modal fusion methods, this paper proposed an end-to-end multi-modal three-dimensional detection network, VAF-Net. First, this paper constructed a noise-resistant virtual point enhancement framework, which utilized images to generate virtual points to densify the point clouds and employed VirConvNet as the backbone network to extract global geometric features while effectively filtering the noise introduced by the virtual points. Second, this paper designed a color semantic feature extraction branch to precisely extract local textures and contextual semantics of images aligned with the three-dimensional space through a semantic lifting module. Finally, this paper proposed a bidirectional cross-attention fusion module based on standard Transformer principles. Through a symmetric dual-stream architecture and a gating mechanism, it achieved deep, stable, and bidirectional interaction between LiDAR geometric features and image semantic features. Experiments on the KITTI three-dimensional object detection benchmark indicate that VAF-Net achieves state-of-the-art performance. Particularly in the detection of hard samples, its 3D average precision reaches 82.74%, significantly outperforming mainstream methods such as Voxel-RCNN. The research results demonstrate that VAF-Net effectively realizes the complementary enhancement of multi-modal information through the standardized bidirectional fusion architecture and significantly improves the detection robustness in complex scenarios.

    参考文献
    相似文献
    引证文献
引用本文

马保全,王韵涵,于寅虎,郝博文,李佳伟,杨洋.虚拟点增强与双向跨注意力融合的3D目标检测[J].哈尔滨工业大学学报,2026,58(8):164. DOI:10.11918/202511097

复制
分享
相关视频

文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2025-11-29
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2026-08-29
  • 出版日期:
文章二维码