Papers

数字孪生合成数据训练路边 LiDAR 检测器夺冠 IEEE BigData 挑战赛

Fixing the Twin, Not the Detector: Sim-to-Real Roadside LiDAR Detection from Digital-Twin Data

Why we picked it

这篇论文的冠军方案没用真实标注,训练只跑在 8 GB 笔记本 GPU 上,靠修正数字孪生的坐标、点密度和标注差异就把逼真度从 0.589 拉到 0.947。做自动驾驶或 roadside 感知的团队可以直接抄它的数据校正流程,成本远低于人工标注。

一个仅用合成数据训练的 CenterPoint 检测器,在 IEEE BigData 2026 Cup 的 UrbanTwin Sim2Real LiDAR 挑战赛中,以 0.5541 的综合得分拿下 LUMPI 赛道第一,并在 V2X-Real 赛道获得第五(0.8102)。全部训练在单块 8 GB 显存的笔记本 GPU 上完成,没有使用任何真实标注。

Body

技术路线的核心是修正数字孪生与真实站点之间的系统性差异,而非升级检测器本身:团队对齐了孪生体的坐标系、点密度与标注框约定,补生成孪生体缺失的行人和骑行者,并加入无标签的街道杂波作为负样本,再按真实尺寸和真实点密度重建孪生物体。一次非学习式的逐单元点密度重采样,就把 LUMPI 赛道的逼真度从 0.589 提升到 0.947,接近真实帧之间的 0.949。

后续的无标签后处理利用固定路边传感器特性:静态结构每帧重复出现,真实物体有特征点统计。逐组件对比验证集与测试集增益后发现,系统性 sim-to-real 修正通常能在测试集上保留四分之三以内的验证增益,而针对验证集少量物体做的拟合选择则不能迁移。

Original · arXiv cs.LG

Fixing the Twin, Not the Detector: Sim-to-Real Roadside LiDAR Detection from Digital-Twin Data

Roadside LiDAR detectors are expensive to train on real labels, and digital twins promise to replace these labels, but a detector trained on a twin meets a different sensor, different objects and different annotation conventions on the real site. We describe the pipeline that placed first in the LUMPI track of the UrbanTwin Sim2Real LiDAR Challenge (IEEE BigData 2026 Cup, 0.5541 combined score) and fifth on its V2X-Real track (0.8102). Every detector is a CenterPoint model trained only on synthetic data, on a single 8 GB laptop GPU. Most of the gain came from correcting differences between the twin and the real site rather than from the detector. We align the twin's coordinates, point density and box convention with the test sensor, generate the pedestrians and riders that the twins lack, plus street clutter as unlabeled negatives, and rebuild twin objects at real size and real point density. A non-learned resampling to the per-cell point density of unlabeled real frames lifts LUMPI realism from 0.589 to 0.947, against 0.949 between real frames. Label-free post-processing then exploits what a fixed roadside sensor guarantees: static structures recur in every frame, and real objects have characteristic point statistics. Reporting the validation and test gain of each component side by side, we find that corrections of systematic sim-to-real differences transfer to the test set, typically keeping a quarter to three quarters of their validation gain, while choices that fit the few objects of a validation set do not.

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include CenterPoint, LiDAR, digital twins, Sim2Real.

Source attribution: curated from arXiv cs.LG (@KuanLin Chen, Feng-jui Kuo).