发表机构
Xidian University（西安电子科技大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对UCF UrbanTwin LUMPI赛道，提出多层级Sim2Real方法，包括数据增强、多检测器集成和真实感优化，最终综合得分0.4692。
AI 中文摘要
我们展示了在第6届DriveX研讨会（ECCV 2026）UCF UrbanTwin Sim2Real LiDAR挑战赛LUMPI赛道上的解决方案。检测器仅使用合成数据进行训练，并在50帧保留的真实LiDAR数据上进行评估；另外提交的50帧合成数据用于评估点云真实感。我们的方法在三个层面解决Sim2Real差距。首先，我们将合成扫描对齐到50k点测试密度，并利用UT-LUMPI几何、基于RangeLDM的采样多样化、稀有类别复制粘贴和面向行人的增强，构建了包含30k条记录的训练池。其次，在相同的纯合成约束下，训练了互补的DSVT检测器和Car/Bus PointPillars专家模型。第三，通过类别感知路由、非对称一致性融合、受限残差召回补充、类别覆盖审计和选择性框尺寸校准来集成预测。真实感分支独立优化，采用径向密度匹配、弱仿射校准和校准集混合。最终提交获得综合得分0.4692，检测得分0.1797，真实感得分0.9035，以及3D mAP@0.5为0.1258。
英文摘要
We present our solution to the LUMPI track of the UCF UrbanTwin Sim2Real LiDAR Challenge at the 6th DriveX Workshop, ECCV 2026. The detector must be trained only on synthetic data and is evaluated on 50 held-out real LiDAR frames; a separate 50-frame synthetic submission is evaluated for point-cloud realism. Our method addresses the Sim2Real gap at three levels. First, we align synthetic scans to the 50k-point test density and build a 30k-record training pool using UT-LUMPI geometry, RangeLDM-based sampling diversification, rare-class copy-paste, and pedestrian-oriented augmentation. Second, complementary DSVT detectors and Car/Bus PointPillars specialists are trained under the same synthetic-only constraint. Third, predictions are integrated by class-aware routing, asymmetric agreement fusion, constrained residual-recall supplementation, class-coverage auditing, and selective box-size calibration. The realism branch is optimized independently with radial-density matching, weak affine calibration, and calibrated set mixing. The final submission obtains a Combined Score of 0.4692, a Detection Score of 0.1797, a Realism Score of 0.9035, and 3D mAP@0.5 of 0.1258.
Comments5 pages,1 figures
