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

TL;DR

This paper introduces SynDRA-BBox, a synthetic dataset for railway object detection, and demonstrates how domain adaptation techniques can transfer knowledge from synthetic to real railway perception tasks.

Contribution

It presents the first synthetic dataset tailored for railway 3D detection and adapts a semi-supervised domain adaptation method for railway perception.

Findings

Synthetic dataset supports railway object detection tasks.

Domain adaptation improves transfer from synthetic to real data.

Promising results validate the approach's effectiveness.

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.