UrbanTwin: Synthetic LiDAR Datasets (LUMPI, V2X-Real-IC, and TUMTraf-I)
This provides a scalable solution for researchers and practitioners in autonomous driving and robotics to enhance training data for LiDAR-based perception tasks, though it is incremental as it builds on existing digital twin and synthetic data methods.
The paper tackles the problem of limited real-world LiDAR data for perception tasks by creating synthetic datasets (UrbanTwin) that replicate three public roadside LiDAR datasets, resulting in high similarity scores and improved 3D object detection performance when models are trained solely on synthetic data and tested on real data.
This article presents UrbanTwin datasets, high-fidelity, realistic replicas of three public roadside lidar datasets: LUMPI, V2X-Real-IC}}, and TUMTraf-I. Each UrbanTwin dataset contains 10K annotated frames corresponding to one of the public datasets. Annotations include 3D bounding boxes, instance segmentation labels, and tracking IDs for six object classes, along with semantic segmentation labels for nine classes. These datasets are synthesized using emulated lidar sensors within realistic digital twins, modeled based on surrounding geometry, road alignment at lane level, and the lane topology and vehicle movement patterns at intersections of the actual locations corresponding to each real dataset. Due to the precise digital twin modeling, the synthetic datasets are well aligned with their real counterparts, offering strong standalone and augmentative value for training deep learning models on tasks such as 3D object detection, tracking, and semantic and instance segmentation. We evaluate the alignment of the synthetic replicas through statistical and structural similarity analysis with real data, and further demonstrate their utility by training 3D object detection models solely on synthetic data and testing them on real, unseen data. The high similarity scores and improved detection performance, compared to the models trained on real data, indicate that the UrbanTwin datasets effectively enhance existing benchmark datasets by increasing sample size and scene diversity. In addition, the digital twins can be adapted to test custom scenarios by modifying the design and dynamics of the simulations. To our knowledge, these are the first digitally synthesized datasets that can replace in-domain real-world datasets for lidar perception tasks. UrbanTwin datasets are publicly available at https://dataverse.harvard.edu/dataverse/ucf-ut.