用拓扑约束提升激光雷达点云生成的逼真度与可解释性
TopoLiDM: Topology-Aware LiDAR Diffusion Models for Interpretable and Realistic LiDAR Point Cloud Generation
- 通过图神经网络与拓扑正则化结合，保留点云全局结构
- 在KITTI-360上实现FRID降低22.6%、MMD降低9.2%
- 生成速度快至1.68样本/秒，适合实际自动驾驶应用
激光雷达场景生成对降低真实数据采集成本、增强自动驾驶感知任务鲁棒性至关重要。现有方法常难以兼顾几何真实感与全局拓扑一致性。近期激光雷达扩散模型（LiDM）虽通过潜空间嵌入提升效率，但限制了对细节几何结构的可解释建模及拓扑一致性保持。为此，我们提出TopoLiDM，将图神经网络与扩散模型结合，并引入拓扑正则化以实现高保真生成。首先训练一个拓扑保持的变分自编码器（VAE），通过图构建和多层图卷积提取潜空间图表示；随后冻结VAE，利用潜扩散模型生成新的拓扑图。引入0维持久同调（PH）约束，确保生成场景符合真实世界的全局拓扑结构。在KITTI-360数据集上的实验表明，TopoLiDM显著优于当前最优方法：Frechet Range Image Distance（FRID）降低22.6%，Minimum Matching Distance（MMD）降低9.2%。模型平均推理速度达1.68样本/秒，具备实际应用扩展性。代码将开源于https://github.com/IRMVLab/TopoLiDM。
原文摘要 · Abstract (English)
LiDAR scene generation is critical for mitigating real-world LiDAR data collection costs and enhancing the robustness of downstream perception tasks in autonomous driving. However, existing methods commonly struggle to capture geometric realism and global topological consistency. Recent LiDAR Diffusion Models (LiDMs) predominantly embed LiDAR points into the latent space for improved generation efficiency, which limits their interpretable ability to model detailed geometric structures and preserve global topological consistency. To address these challenges, we propose TopoLiDM, a novel framework that integrates graph neural networks (GNNs) with diffusion models under topological regularization for high-fidelity LiDAR generation. Our approach first trains a topological-preserving VAE to extract latent graph representations by graph construction and multiple graph convolutional layers. Then we freeze the VAE and generate novel latent topological graphs through the latent diffusion models. We also introduce 0-dimensional persistent homology (PH) constraints, ensuring the generated LiDAR scenes adhere to real-world global topological structures. Extensive experiments on the KITTI-360 dataset demonstrate TopoLiDM's superiority over state-of-the-art methods, achieving improvements of 22.6% lower Frechet Range Image Distance (FRID) and 9.2% lower Minimum Matching Distance (MMD). Notably, our model also enables fast generation speed with an average inference time of 1.68 samples/s, showcasing its scalability for real-world applications. We will release the related codes at https://github.com/IRMVLab/TopoLiDM.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。