论文/arXiv 2609.24825·1 个月前

ZVeC：用于实例级车辆提取与生成式点云补全的零样本框架

ZVeC将地下场景点云补全分解为车辆实例重建，用条件扩散模型零样本补全，并在真实地下停车LiDAR基准验证。

AI 整理

• 感知与导航
• 数据与采集
• 评测与安全

原文链接：ZVeC: A Zero-Shot Framework for Instance-Level Vehicle Extraction and Generative Point Cloud Completion（arXiv 2609.24825 · 发布日期：2026-09-21）

ZVeC: A Zero-Shot Framework for Instance-Level Vehicle Extraction and Generative Point Cloud Completion

ZVeC将地下场景点云补全分解为车辆实例重建，用条件扩散模型零样本补全，并在真实地下停车LiDAR基准验证。

核心要点

• 地下LiDAR点云因遮挡和视角有限严重不完整，且缺少大规模监督数据。
• ZVeC将场景分解为语义对象实例，逐个补全车辆后再重组回原场景。
• 每个车辆由深度和3D Gaussian条件扩散模型补全，以利用泛化几何先验。
• 实验在真实地下停车稠密LiDAR基准上，相比场景级基线在指标和视觉质量上持续改进。

验证范围与局限：验证范围限于真实地下停车环境稠密LiDAR基准中的车辆点云补全。

英文摘要 (Abstract)

LiDAR point clouds acquired in underground environments exhibit severe geometric incompleteness due to occlusions and limited sensor viewpoints, making reliable point cloud completion challenging without large supervised datasets. We propose ZVeC, a zero-shot, instance-driven framework that reformulates scene-level completion as compositional object-level reconstruction. By decomposing a scene into semantic object instances, ZVeC reduces reconstruction ambiguity in cluttered environments while eliminating the need for scenario-specific training. Each segmented vehicle is completed independently using a depth- and 3D Gaussian-conditioned diffusion model that exploits generalized geometric priors before the reconstructed instances are recomposed into the original scene. To evaluate our approach, we construct a real-world dense LiDAR benchmark of underground parking environments. Experimental results demonstrate consistent improvements over representative scene-level baselines in both quantitative metrics and visual quality. The completed point cloud differs substantially from the measured input (average KL divergence ~ 2.1), yet reducing the input to only 1% of the original LiDAR measurements changes the completed reconstruction only marginally (KL divergence < 0.50). This demonstrates that ZVeC produces geometrically consistent completions even under extreme input sparsity.

作者：Daisy Li, Kyle Gao, Quanyun Wu, B. Jutzi, John Zelek, Jonathan Li

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2609.24825：ZVeC: A Zero-Shot Framework for Instance-Level Vehicle Extraction and Generative Point Cloud Completion访问 2026-10-21