BikeScenes: Online LiDAR Semantic Segmentation for Bicycles

TL;DR

This paper presents BikeScenes, a new LiDAR dataset and segmentation approach tailored for bicycles, demonstrating improved performance through domain-specific training and addressing unique challenges of bicycle-mounted perception systems.

Contribution

Introduction of the BikeScenes-lidarseg dataset and a tailored LiDAR segmentation method for bicycles, bridging the automotive-to-bicycle domain gap.

Findings

Fine-tuning on BikeScenes yields 63.6% mIoU.

Pre-training on SemanticKITTI alone achieves 13.8% mIoU.

Domain-specific training significantly improves segmentation performance.

Abstract

The vulnerability of cyclists, exacerbated by the rising popularity of faster e-bikes, motivates adapting automotive perception technologies for bicycle safety. We use our multi-sensor 'SenseBike' research platform to develop and evaluate a 3D LiDAR segmentation approach tailored to bicycles. To bridge the automotive-to-bicycle domain gap, we introduce the novel BikeScenes-lidarseg Dataset, comprising 3021 consecutive LiDAR scans around the university campus of the TU Delft, semantically annotated for 29 dynamic and static classes. By evaluating model performance, we demonstrate that fine-tuning on our BikeScenes dataset achieves a mean Intersection-over-Union (mIoU) of 63.6%, significantly outperforming the 13.8% obtained with SemanticKITTI pre-training alone. This result underscores the necessity and effectiveness of domain-specific training. We highlight key challenges specific to bicycle-mounted, hardware-constrained perception systems and contribute the BikeScenes dataset as a resource for advancing research in cyclist-centric LiDAR segmentation.