TY - JOUR A1 - Bieshaar, Maarten A1 - Zernetsch, Stefan A1 - Hubert, Andreas A1 - Sick, Bernhard A1 - Doll, Konrad T1 - Cooperative Starting Movement Detection of Cyclists Using Convolutional Neural Networks and a Boosted Stacking Ensemble JF - IEEE Transactions on Intelligent Vehicles N2 - In the future, vehicles and other traffic participants will be interconnected and equipped with various types of sensors, allowing for cooperation on different levels, such as situation prediction or intention detection. In this paper, we present a cooperative approach for starting movement detection of cyclists using a boosted stacking ensemble approach realizing feature- and decision-level cooperation. We introduce a novel method based on a three-dimensional convolutional neural network (CNN) to detect starting motions on image sequences by learning spatio-temporal features. The CNN is complemented by a smart device based starting movement detection originating from smart devices carried by the cyclist. Both model outputs are combined in a stacking ensemble approach using an extreme gradient boosting classifier resulting in a fast and yet robust cooperative starting movement detector. We evaluate our cooperative approach on real-world data originating from experiments with 49 test subjects consisting of 84 starting motions. KW - Fahrerassistenzsystem KW - Fahrrad KW - Sensortechnik Y1 - 2018 VL - 3 IS - 4 SP - 534 EP - 544 ER - TY - CHAP A1 - Eilbrecht, Jan A1 - Bieshaar, Maarten A1 - Zernetsch, Stefan A1 - Doll, Konrad A1 - Sick, Bernhard A1 - Stursberg, Olaf T1 - Model-predictive planning for autonomous vehicles anticipating intentions of vulnerable road users by artificial neural networks T2 - 2017 IEEE Symposium Series on Computational Intelligence (SSCI) KW - Fahrerassistenzsystem KW - Fußgänger KW - Radfahrer KW - Autonomes Fahrzeug Y1 - 2017 U6 - https://doi.org/10.1109/SSCI.2017.8285249 SP - 1 EP - 8 PB - IEEE CY - Honolulu, HI, USA ER - TY - JOUR A1 - Goldhammer, Michael A1 - Köhler, Sebastian A1 - Zernetsch, Stefan A1 - Doll, Konrad A1 - Sick, Bernhard A1 - Dietmayer, Klaus T1 - Intentions of Vulnerable Road Users – Detection and Forecasting by Means of Machine Learning JF - IEEE Transactions on Intelligent Transportation Systems KW - Fahrerassistenzsystem KW - Fußgänger KW - Radfahrer Y1 - 2019 UR - https://doi.org/10.1109/TITS.2019.2923319 U6 - https://doi.org/10.1109/TITS.2019.2923319 VL - 2020 IS - 21 / 7 SP - 3035 EP - 3045 ER - TY - CHAP A1 - Bieshaar, Maarten A1 - Zernetsch, Stefan A1 - Depping, Malte A1 - Sick, Bernhard A1 - Doll, Konrad T1 - Cooperative Starting Intention Detection of Cyclists based on Smart Devices and Infrastructure T2 - 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC) KW - Fahrerassistenzsystem KW - Radfahrer Y1 - 2017 U6 - https://doi.org/10.1109/ITSC.2017.8317691 SP - 1 EP - 8 PB - IEEE CY - Yokohama, Japan ER - TY - CHAP A1 - Bieshaar, Maarten A1 - Reitberger, Günther A1 - Kreß, Viktor A1 - Zernetsch, Stefan A1 - Doll, Konrad A1 - Fuchs, Erich A1 - Sick, Bernhard T1 - Highly Automated Learning for Improved Active Safety of Vulnerable Road Users T2 - ACM Chapters Computer Science in Cars Symposium (CSCS-17) KW - Fahrerassistenzsystem Y1 - 2017 VL - 2017 CY - München ER - TY - CHAP A1 - Kress, Viktor A1 - Schreck, Steven A1 - Zernetsch, Stefan A1 - Doll, Konrad A1 - Sick, Bernhard T1 - Pose Based Action Recognition of Vulnerable Road Users Using Recurrent Neural Networks T2 - 2020 IEEE Symposium Series on Computational Intelligence (SSCI) N2 - This work investigates the use of knowledge about three dimensional (3D) poses and Recurrent Neural Networks (RNNs) for detection of basic movements, such as wait, start, move, stop, turn left, turn right, and no turn, of pedestrians and cyclists in road traffic. The 3D poses model the posture of individual body parts of these vulnerable road users (VRUs). Fields of application for this technology are, for example, driver assistance systems or autonomous driving functions of vehicles. In road traffic, VRUs are often occluded and only become visible in the immediate vicinity of the vehicle. Hence, our proposed approach is able to classify basic movements after different and especially short observation periods. The classification will then be successively improved in case of a longer observation. This allows countermeasures, such as emergency braking, to be initiated early if necessary. The benefits of using 3D poses are evaluated by a comparison with a method based solely on the head trajectory. We also investigate the effects of different observation periods. Overall, knowledge about 3D poses improves the basic movement detection, in particular for short observation periods. The greatest improvements are achieved for the basic movements start, stop, turn left, and turn right. KW - Fahrerassistenzsystem KW - Autonomes Fahrzeug KW - Fußgänger KW - Fahrrad Y1 - 2020 SN - 978-1-7281-2548-0 U6 - https://doi.org/10.1109/SSCI47803.2020.9308462 SP - 2723 EP - 2730 PB - IEEE CY - Canberra, Australia ER - TY - CHAP A1 - Zernetsch, Stefan A1 - Schreck, Steven A1 - Kress, Viktor A1 - Doll, Konrad A1 - Sick, Bernhard T1 - Image Sequence Based Cyclist Action Recognition Using Multi-Stream 3D Convolution T2 - 2020 25th International Conference on Pattern Recognition (ICPR) N2 - In this article, we present an approach to detect basic movements of cyclists in real world traffic situations based on image sequences, optical flow (OF) sequences, and past positions using a multi-stream 3D convolutional neural network (3D-ConvNet) architecture. To resolve occlusions of cyclists by other traffic participants or road structures, we use a wide angle stereo camera system mounted at a heavily frequented public intersection. We created a large dataset consisting of 1,639 video sequences containing cyclists, recorded in real world traffic, resulting in over 1.1 million samples. Through modeling the cyclists' behavior by a state machine of basic cyclist movements, our approach takes every situation into account and is not limited to certain scenarios. We compare our method to an approach solely based on position sequences. Both methods are evaluated taking into account frame wise and scene wise classification results of basic movements, and detection times of basic movement transitions, where our approach outperforms the position based approach by producing more reliable detections with shorter detection times. Our code and parts of our dataset are made publicly available. KW - Fahrerassistenzsystem KW - Radfahrer Y1 - 2020 UR - https://ieeexplore.ieee.org/document/9413233 ER - TY - CHAP A1 - Kreß, Viktor A1 - Zernetsch, Stefan A1 - Doll, Konrad A1 - Sick, Bernhard T1 - Pose Based Trajectory Forecast of Vulnerable Road Users Using Recurrent Neural Networks T2 - ICPR 2021: Pattern Recognition. ICPR International Workshops and Challenges N2 - In this work, we use Recurrent Neural Networks (RNNs) in form of Gated Recurrent Unit (GRU) networks to forecast trajectories of vulnerable road users (VRUs), such as pedestrians and cyclists, in road traffic utilizing the past trajectory and 3D poses as input. The 3D poses represent the postures and movements of limbs and torso and contain early indicators for the transition between motion types, e.g. wait, start, move, and stop. VRUs often only become visible from the perspective of an approaching vehicle shortly before dangerous situations occur. Therefore, a network architecture is required which is able to forecast trajectories after short time periods and is able to improve the forecasts in case of longer observations. This motivates us to use GRU networks, which are able to use time series of varying duration as inputs, and to investigate the effects of different observation periods on the forecasting results. Our approach is able to make reasonable forecasts even for short observation periods. The use of poses improves the forecasting accuracy, especially for short observation periods compared to a solely head trajectory based approach. Different motion types benefit to different extent from the use of poses and longer observation periods. KW - 3D human trajectory forecast KW - Vulnerable road users KW - 3D body pose KW - Fahrerassistenzsystem KW - Fußgänger KW - Radfahrer Y1 - 2020 SN - 978-3-030-68763-2 U6 - https://doi.org/https://doi.org/10.1007/978-3-030-68763-2_5 SP - 57 EP - 71 PB - Springer International Publishing CY - Cham ER - TY - THES A1 - Zernetsch, Stefan T1 - Maschinelle Lernverfahren zur videobasierten Intentionserkennung von Radfahrern mit stationären Kameras KW - Maschinelles Lernen KW - Autonomes Fahrzeug KW - Fahrrad KW - Videoüberwachung Y1 - 2022 UR - https://kobra.uni-kassel.de/handle/123456789/13966 SN - 978-3-7376-1053-7 ER - TY - CHAP A1 - Hubert, Andreas A1 - Zernetsch, Stefan A1 - Doll, Konrad A1 - Sick, Bernhard T1 - Cyclists starting behavior at intersections T2 - 2017 IEEE Intelligent Vehicles Symposium (IV) KW - Fahrerassistenzsystem KW - Radfahrer Y1 - 2017 U6 - https://doi.org/10.1109/IVS.2017.7995856 SP - 1071 EP - 1077 PB - IEEE CY - Los Angeles, CA, USA ER -