Fabian B. Flohr
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Fabian B. Flohr has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Dariu M. Gavrila
Intelligent Vehicles Group, TU Delft, Delft, The Netherlands
2 shared publications - Erik Schuetz
Munich University of Applied Sciences
2 shared publications - Kanak Mazumder
Munich University of Applied Sciences
2 shared publications - Max A. Buettner
Munich University of Applied Sciences
2 shared publications - Alexander von Hoffmann
Elektrotechnik Feinwerktechnik Informationstechnik, Technische Hochschule Nürnberg Georg Simon Ohm, Keßlerplatz 12, 90489 Nuremberg, Germany
1 shared publication - Constantin Selzer
Munich University of Applied Sciences
1 shared publication
Latest Publications
- 3 August 2026
- Published by Springer Nature
- p. 212-226
BikeActions: An Open Platform and Benchmark for Cyclist-Centric VRU Action Recognition
- 3 August 2026
- Published by Springer Nature
- p. 212-226
BikeActions: An Open Platform and Benchmark for Cyclist-Centric VRU Action Recognition
- 3 August 2026
- Published by Springer Nature
- p. 198-211
SatMap: Revisiting Satellite Maps as Prior for Online HD Map Construction
- 3 August 2026
- Published by Springer Nature
- p. 198-211
SatMap: Revisiting Satellite Maps as Prior for Online HD Map Construction
- 9 July 2026
- Published by Springer Nature
- p. 151-167
Beyond the Dashboard: A Collaborative UI Framework for Autonomous Systems Engineering and Design Research
- 9 July 2026
- Published by Springer Nature
- p. 151-167
Beyond the Dashboard: A Collaborative UI Framework for Autonomous Systems Engineering and Design Research
- Open Access
HABIT: Human Action Benchmark for Interactive Traffic in CARLA
Current autonomous driving (AD) simulations are critically limited by their inadequate representation of realistic and diverse human behavior, which is essential for ensuring safety and reliability. Existing benchmarks often simplify pedestrian interactions, failing to capture complex, dynamic intentions and varied responses critical for robust system deployment. To overcome this, we introduce HABIT (Human Action Benchmark for Interactive Traffic), a high-fidelity simulation benchmark. HABIT integrates real-world human motion, sourced from mocap and videos, into CARLA (Car Learning to Act, a full autonomous driving simulator) via a modular, extensible, and physically consistent motion retargeting pipeline. From an initial pool of approximately 30,000 retargeted motions, we curate 4,730 traffic-compatible pedestrian motions, standardized in SMPL format for physically consistent trajectories. HABIT seamlessly integrates with CARLA's Leaderboard, enabling automated scenario generation and rigorous agent evaluation. Our safety metrics, including Abbreviated Injury Scale (AIS) and False Positive Braking Rate (FPBR), reveal critical failure modes in state-of-the-art AD agents missed by prior evaluations. Evaluating three state-of-the-art autonomous driving agents, InterFuser, TransFuser, and BEVDriver, demonstrates how HABIT exposes planner weaknesses that remain hidden in scripted simulations. Despite achieving close or equal to zero collisions per kilometer on the CARLA Leaderboard, the autonomous agents perform notably worse on HABIT, with up to 7.43 collisions/km and a 12.94% AIS 3+ injury risk, and they brake unnecessarily in up to 33% of cases. All components are publicly released to support reproducible, pedestrian-aware AI research.
Open AccessHABIT: Human Action Benchmark for Interactive Traffic in CARLA
Current autonomous driving (AD) simulations are critically limited by their inadequate representation of realistic and diverse human behavior, which is essential for ensuring safety and reliability. Existing benchmarks often simplify pedestrian interactions, failing to capture complex, dynamic intentions and varied responses critical for robust system deployment. To overcome this, we introduce HABIT (Human Action Benchmark for Interactive Traffic), a high-fidelity simulation benchmark. HABIT integrates real-world human motion, sourced from mocap and videos, into CARLA (Car Learning to Act, a full autonomous driving simulator) via a modular, extensible, and physically consistent motion retargeting pipeline. From an initial pool of approximately 30,000 retargeted motions, we curate 4,730 traffic-compatible pedestrian motions, standardized in SMPL format for physically consistent trajectories. HABIT seamlessly integrates with CARLA's Leaderboard, enabling automated scenario generation and rigorous agent evaluation. Our safety metrics, including Abbreviated Injury Scale (AIS) and False Positive Braking Rate (FPBR), reveal critical failure modes in state-of-the-art AD agents missed by prior evaluations. Evaluating three state-of-the-art autonomous driving agents, InterFuser, TransFuser, and BEVDriver, demonstrates how HABIT exposes planner weaknesses that remain hidden in scripted simulations. Despite achieving close or equal to zero collisions per kilometer on the CARLA Leaderboard, the autonomous agents perform notably worse on HABIT, with up to 7.43 collisions/km and a 12.94% AIS 3+ injury risk, and they brake unnecessarily in up to 33% of cases. All components are publicly released to support reproducible, pedestrian-aware AI research.
- Open Access
A Review of Trajectory Prediction Methods for the Vulnerable Road User
Predicting the trajectory of other road users, especially vulnerable road users (VRUs), is an important aspect of safety and planning efficiency for autonomous vehicles. With recent advances in Deep-Learning-based approaches in this field, physics- and classical Machine-Learning-based methods cannot exhibit competitive results compared to the former. Hence, this paper provides an extensive review of recent Deep-Learning-based methods in trajectory prediction for VRUs and autonomous driving in general. We review the state and context representations and architectural insights of selected methods, divided into categories according to their primary prediction scheme. Additionally, we summarize reported results on popular datasets for all methods presented in this review. The results show that conditional variational autoencoders achieve the best overall results on both pedestrian and autonomous driving datasets. Finally, we outline possible future research directions for the field of trajectory prediction in autonomous driving.
Open AccessA Review of Trajectory Prediction Methods for the Vulnerable Road User
Predicting the trajectory of other road users, especially vulnerable road users (VRUs), is an important aspect of safety and planning efficiency for autonomous vehicles. With recent advances in Deep-Learning-based approaches in this field, physics- and classical Machine-Learning-based methods cannot exhibit competitive results compared to the former. Hence, this paper provides an extensive review of recent Deep-Learning-based methods in trajectory prediction for VRUs and autonomous driving in general. We review the state and context representations and architectural insights of selected methods, divided into categories according to their primary prediction scheme. Additionally, we summarize reported results on popular datasets for all methods presented in this review. The results show that conditional variational autoencoders achieve the best overall results on both pedestrian and autonomous driving datasets. Finally, we outline possible future research directions for the field of trajectory prediction in autonomous driving.
- Open Access
- 18 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 323-326
Using Node-RED as a Low-Code Approach to Model Interaction Logic of Machine-Learning-Supported eHMIs for the Virtual Driving Simulator Carla
- 18 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 323-326
Open AccessUsing Node-RED as a Low-Code Approach to Model Interaction Logic of Machine-Learning-Supported eHMIs for the Virtual Driving Simulator Carla
- 5 February 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 41 (8) , 1844-1861
EuroCity Persons: A Novel Benchmark for Person Detection in Traffic Scenes
Big data has had a great share in the success of deep learning in computer vision. Recent works suggest that there is significant further potential to increase object detection performance by utilizing even bigger datasets. In this paper, we introduce the EuroCity Persons dataset, which provides a large number of highly diverse, accurate and detailed annotations of pedestrians, cyclists and other riders in urban traffic scenes. The images for this dataset were collected on-board a moving vehicle in 31 cities of 12 European countries. With over 238200 person instances manually labeled in over 47300 images, EuroCity Persons is nearly one order of magnitude larger than datasets used previously for person detection in traffic scenes. The dataset furthermore contains a large number of person orientation annotations (over 211200). We optimize four state-of-the-art deep learning approaches (Faster R-CNN, R-FCN, SSD and YOLOv3) to serve as baselines for the new object detection benchmark. We analyze the generalization capabilities of these detectors when trained with the new dataset. We furthermore study the effect of the training set size, the dataset diversity (day- vs. night-time, geographical region), the dataset detail (i.e. availability of object orientation information) and the annotation quality on the detector performance. Finally, we analyze error sources and discuss the road ahead.
- 5 February 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 41 (8) , 1844-1861
EuroCity Persons: A Novel Benchmark for Person Detection in Traffic Scenes
Big data has had a great share in the success of deep learning in computer vision. Recent works suggest that there is significant further potential to increase object detection performance by utilizing even bigger datasets. In this paper, we introduce the EuroCity Persons dataset, which provides a large number of highly diverse, accurate and detailed annotations of pedestrians, cyclists and other riders in urban traffic scenes. The images for this dataset were collected on-board a moving vehicle in 31 cities of 12 European countries. With over 238200 person instances manually labeled in over 47300 images, EuroCity Persons is nearly one order of magnitude larger than datasets used previously for person detection in traffic scenes. The dataset furthermore contains a large number of person orientation annotations (over 211200). We optimize four state-of-the-art deep learning approaches (Faster R-CNN, R-FCN, SSD and YOLOv3) to serve as baselines for the new object detection benchmark. We analyze the generalization capabilities of these detectors when trained with the new dataset. We furthermore study the effect of the training set size, the dataset diversity (day- vs. night-time, geographical region), the dataset detail (i.e. availability of object orientation information) and the annotation quality on the detector performance. Finally, we analyze error sources and discuss the road ahead.
- Open Access
- 2 July 2018
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 127 (3) , 239-262
Context-Based Path Prediction for Targets with Switching Dynamics
- 2 July 2018
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 127 (3) , 239-262
Open AccessContext-Based Path Prediction for Targets with Switching Dynamics