
AD
Antong Deng, Yuxin Zhang, Zhongxiang Xing, Zhenzhen Lin, Jiong Mu
· 1 min read
ResearcharXiv cs.CV
Third-Place Solution for the UCF UrbanTwin V2X-Real Track at ECCV 2026: Synthetic Point Cloud Generation and Sim-to-Real 3D Object Detection
arXiv:2611.18778v1 Announce Type: new
Abstract: This report presents a complete technical solution for the UrbanTwin V2X-Real Sim2Real LiDAR Challenge, covering both synthetic point cloud generation and 3D object detection. For data generation, we adopt per-axis quantile calibration. Public V2X-Real training data are used to construct the real reference distribution, and independent one-dimensional quantile mappings are estimated for the x, y, and z axes of the base synthetic point clouds. The calibrated point clouds are then resampled using a fixed random seed and clipped to the valid spatial range, producing 50 frames with 2,000 float32 XYZ points per frame. For detection, we use a two-stage Voxel R-CNN (VoxelRCNN) detector trained with RangeLDM-enhanced synthetic point-cloud data, point clouds downsampled to 50k points per frame, and reconstructed pedestrian data. Training is performed on four NVIDIA GeForce RTX 4090 GPUs with a global batch size of 16 for 40 epochs. Final testing uses direct single-model inference with the original VoxelRCNN model. On the official evaluation server, the normalized Realism score is 0.7905 and the normalized Detection score is 0.5583, with mAP@0.5 of 0.3908, BEV mAP@0.5 of 0.4219, and mAP@0.7 of 0.1047. According to the official weighting rule, the final combined score is 0.6 x 0.5583 + 0.4 x 0.7905 = 0.6512. This report documents the complete implementation details of data construction, model training, inference, submission formatting, and official evaluation.
Original source
This story was published by arXiv cs.CV and written by Antong Deng, Yuxin Zhang, Zhongxiang Xing, Zhenzhen Lin, Jiong Mu. SyncAI.news shows a preview; the complete article is on the publisher's site.
Read the full story on arxiv.org


