Research

bikeActions

PaperPublisherCodeDataset
Six panels: three urban street camera images from a bicycle with 2D skeletons and boxes on a signalling cyclist, a crossing pedestrian and a cyclist in a narrow lane, above the matching colour-coded LiDAR depth images.

Cyclists signal with their bodies — an arm out, a shoulder check, a shift in balance. bikeActions labels those cues on real recordings so models can learn them.

What it does

Three of the dataset’s authors started on it as students in the lab. The recordings come from the instrumented research bicycle and the test vehicle; the labels cover the manoeuvres that matter for an approaching car.

Abstract

Anticipating the intentions of Vulnerable Road Users (VRUs) is a critical challenge for safe autonomous driving (AD) and mobile robotics. While current research predominantly focuses on pedestrian crossing behaviors from a vehicle’s perspective, interactions within dense shared spaces remain underexplored. To bridge this gap, we introduce FUSE-Bike, the first fully open perception platform of its kind. Equipped with two LiDARs, a camera, and GNSS, it facilitates high-fidelity, close-range data capture directly from a cyclist’s viewpoint. Leveraging this platform, we present BikeActions, a novel multi-modal dataset comprising 852 annotated samples across 5 distinct action classes, specifically tailored to improve VRU behavior modeling. We establish a rigorous benchmark by evaluating state-of-the-art graph convolution and transformer-based models on our publicly released data splits, establishing the first performance baselines for this challenging task. We release the full dataset together with data curation tools, the open hardware design, and the benchmark code to foster future research in VRU action understanding under https://iv.ee.hm.edu/bikeactions/.

Result

A public dataset and benchmark for cyclist action classification, recorded with the lab's instrumented bicycle and research vehicle.

Citation

@inproceedings{Buettner2026bikeActions,
title = {bikeActions: Action Classification Dataset for Behavior Understanding of Vulnerable Road Users},
author = {Max A. Buettner and Kanak Mazumder and Luca Koecher and Mario Finkbeiner and Sebastian Niebler and Fabian B. Flohr},
url = {https://arxiv.org/abs/2601.10521},
year  = {2026},
date = {2026-01-01},
urldate = {2026-01-01},
booktitle = {International Conference on Pattern Recognition (ICPR)},
publisher = {IEEE},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}