A Data-Centric Approach to 3D Semantic Segmentation of Railway Scenes
This work addresses railway-specific perception challenges for autonomous trains, representing an incremental domain-specific advancement.
The paper tackled the problem of LiDAR-based semantic segmentation for autonomous trains by introducing two data augmentation methods (person instance pasting and track sparsification) to improve performance on distant objects in railway scenes, resulting in significant improvements in distant-range segmentation while maintaining close-range accuracy on the OSDaR23 dataset.
LiDAR-based semantic segmentation is critical for autonomous trains, requiring accurate predictions across varying distances. This paper introduces two targeted data augmentation methods designed to improve segmentation performance on the railway-specific OSDaR23 dataset. The person instance pasting method enhances segmentation of pedestrians at distant ranges by injecting realistic variations into the dataset. The track sparsification method redistributes point density in LiDAR scans, improving track segmentation at far distances with minimal impact on close-range accuracy. Both methods are evaluated using a state-of-the-art 3D semantic segmentation network, demonstrating significant improvements in distant-range performance while maintaining robustness in close-range predictions. We establish the first 3D semantic segmentation benchmark for OSDaR23, demonstrating the potential of data-centric approaches to address railway-specific challenges in autonomous train perception.