Towards Railway Domain Adaptation for LiDAR-based 3D Detection: Road-to-Rail and Sim-to-Real via SynDRA-BBox

2025-07-22

AI Summary

The railway domain has long suffered from a lack of publicly available, realistic, and accurately annotated 3D perception datasets, hindering rigorous validation and real-world deployment of LiDAR-based 3D object detection algorithms. Method: We introduce SynDRA-BBox—the first synthetic 3D detection dataset specifically designed for railway scenarios—featuring physically consistent LiDAR point cloud simulation and the first domain-specific joint 2D/3D annotation. To bridge the domain gaps between simulation and reality, and between automotive and railway environments, we propose a semi-supervised domain adaptation framework enabling cross-platform knowledge transfer. Results: Experiments demonstrate that models trained on SynDRA-BBox achieve significantly improved 3D detection mAP on real-world railway LiDAR data, outperforming those trained on general-purpose synthetic datasets in cross-domain generalization. This work advances algorithm development and trustworthy evaluation for intelligent railway perception.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: 3D Computer Vision

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Web applications in cross-disciplinary domains and verticals such as mixed reality, smart cities, and digital health

Abstract

In recent years, interest in automatic train operations has significantly increased. To enable advanced functionalities, robust vision-based algorithms are essential for perceiving and understanding the surrounding environment. However, the railway sector suffers from a lack of publicly available real-world annotated datasets, making it challenging to test and validate new perception solutions in this domain. To address this gap, we introduce SynDRA-BBox, a synthetic dataset designed to support object detection and other vision-based tasks in realistic railway scenarios. To the best of our knowledge, is the first synthetic dataset specifically tailored for 2D and 3D object detection in the railway domain, the dataset is publicly available at https://syndra.retis.santannapisa.it. In the presented evaluation, a state-of-the-art semi-supervised domain adaptation method, originally developed for automotive perception, is adapted to the railway context, enabling the transferability of synthetic data to 3D object detection. Experimental results demonstrate promising performance, highlighting the effectiveness of synthetic datasets and domain adaptation techniques in advancing perception capabilities for railway environments.

Problem

• Lack of real-world annotated railway datasets for perception testing
• Need synthetic data for 2D/3D object detection in railways
• Adapting automotive domain methods to railway environments

Innovation

• Synthetic dataset SynDRA-BBox for railway
• Semi-supervised domain adaptation method
• Transfer synthetic data to 3D detection