no code implementations • 30 Apr 2024 • Marlon Steiner, Marvin Klemp, Christoph Stiller
This paper addresses that gap by introducing a novel approach to motion prediction, focusing on predicting agent-pair covariance matrices in a ``scene-centric'' manner, which can then be used to model Gaussian joint PDFs for all agent-pairs in a scene.
no code implementations • 18 Mar 2024 • Johannes Fischer, Kevin Rösch, Martin Lauer, Christoph Stiller
To resolve this issue, we propose the novel Physics-Informed Trajectory Autoencoder (PITA) architecture, whichincorporates a physical dynamics model into the loss functionof the autoencoder.
no code implementations • 1 Feb 2024 • Annika Meyer, Christoph Stiller
This representation can be used to describe lane borders, markings, but also implicit features such as centerlines of lanes.
2 code implementations • 12 Feb 2023 • Royden Wagner, Carlos Fernandez Lopez, Christoph Stiller
Self-supervised learning, which is strikingly referred to as the dark matter of intelligence, is gaining more attention in biomedical applications of deep learning.
1 code implementation • ICCV 2023 • Annika Hagemann, Moritz Knorr, Christoph Stiller
The self-calibrating bundle adjustment layer optimizes camera intrinsics through classical Gauss-Newton steps and can be adapted to different camera models without re-training.
no code implementations • 17 Oct 2022 • Kevin Rösch, Florian Heidecker, Julian Truetsch, Kamil Kowol, Clemens Schicktanz, Maarten Bieshaar, Bernhard Sick, Christoph Stiller
Based on these predictions - and additional contextual information such as the course of the road, (traffic) rules, and interaction with other road users - the highly automated vehicle (HAV) must be able to reliably and safely perform the task assigned to it, e. g., moving from point A to B.
no code implementations • 28 Jul 2022 • Miguel Ángel Muñoz-Bañón, Jan-Hendrik Pauls, Haohao Hu, Christoph Stiller, Francisco A. Candelas, Fernando Torres
Localization in aerial imagery-based maps offers many advantages, such as global consistency, geo-referenced maps, and the availability of publicly accessible data.
no code implementations • 22 Jul 2022 • Timo Sämann, Ahmed Mostafa Hammam, Andrei Bursuc, Christoph Stiller, Horst-Michael Groß
Albeit effective, only few works haveimproved the understanding and the performance of weight averaging. Here, we revisit this approach and show that a simple weight fusion (WF)strategy can lead to a significantly improved predictive performance andcalibration.
no code implementations • 19 Apr 2022 • Sven Richter, Frank Bieder, Sascha Wirges, Christoph Stiller
We present a new method to combine evidential top-view grid maps estimated based on heterogeneous sensor sources.
no code implementations • 16 Apr 2022 • Sven Richter, Frank Bieder, Sascha Wirges, Christoph Stiller
We present a generic evidential grid mapping pipeline designed for imaging sensors such as LiDARs and cameras.
1 code implementation • 2 Mar 2022 • Sascha Wirges, Kevin Rösch, Frank Bieder, Christoph Stiller
We propose a fast and robust method to estimate the ground surface from LIDAR measurements on an automated vehicle.
no code implementations • 2 Mar 2022 • Frank Bieder, Maximilian Link, Simon Romanski, Haohao Hu, Christoph Stiller
In particular, we fuse learned features from complementary representations.
no code implementations • 28 Feb 2022 • Haohao Hu, Fengze Han, Frank Bieder, Jan-Hendrik Pauls, Christoph Stiller
To calibrate the stereo camera, a photometric error function is builded and the LiDAR depth is involved to transform key points from one camera to another.
no code implementations • 28 Feb 2022 • Haohao Hu, Hexing Yang, Jian Wu, Xiao Lei, Frank Bieder, Jan-Hendrik Pauls, Christoph Stiller
Since a 3D surface can be usually observed from multiple camera images with different view poses, an optimal image patch selection for the texturing and an optimal semantic class estimation for the semantic mapping are still challenging.
no code implementations • 14 Oct 2021 • Annika Hagemann, Moritz Knorr, Christoph Stiller
We demonstrate the effectiveness of modeling dynamic deformations using different calibration targets and show its significance in a structure-from-motion application.
no code implementations • 18 Aug 2021 • Lukas Stäcker, Juncong Fei, Philipp Heidenreich, Frank Bonarens, Jason Rambach, Didier Stricker, Christoph Stiller
We therefore perform a case study of the deployment of two representative object detection networks on an edge AI platform.
no code implementations • 28 Jul 2021 • Annika Hagemann, Moritz Knorr, Holger Janssen, Christoph Stiller
Accurate camera calibration is a precondition for many computer vision applications.
no code implementations • 15 Jul 2021 • Danial Kamran, Tizian Engelgeh, Marvin Busch, Johannes Fischer, Christoph Stiller
Despite recent advances in reinforcement learning (RL), its application in safety critical domains like autonomous vehicles is still challenging.
Autonomous Vehicles Distributional Reinforcement Learning +2
1 code implementation • 1 Jul 2021 • Kunyu Peng, Juncong Fei, Kailun Yang, Alina Roitberg, Jiaming Zhang, Frank Bieder, Philipp Heidenreich, Christoph Stiller, Rainer Stiefelhagen
At the heart of all automated driving systems is the ability to sense the surroundings, e. g., through semantic segmentation of LiDAR sequences, which experienced a remarkable progress due to the release of large datasets such as SemanticKITTI and nuScenes-LidarSeg.
no code implementations • 10 May 2021 • Juncong Fei, Kunyu Peng, Philipp Heidenreich, Frank Bieder, Christoph Stiller
The recent publication of the SemanticKITTI dataset stimulates the research on semantic segmentation of LiDAR point clouds in urban scenarios.
1 code implementation • 26 Mar 2021 • Annika Meyer, Philipp Skudlik, Jan-Hendrik Pauls, Christoph Stiller
Reformulating the problem of polyline detection as a bottom-up composition of small line segments allows to detect bounded, dashed and continuous polylines with a single head.
no code implementations • 5 Mar 2021 • Florian Heidecker, Jasmin Breitenstein, Kevin Rösch, Jonas Löhdefink, Maarten Bieshaar, Christoph Stiller, Tim Fingscheidt, Bernhard Sick
Systems and functions that rely on machine learning (ML) are the basis of highly automated driving.
no code implementations • 25 Sep 2020 • Juncong Fei, Wenbo Chen, Philipp Heidenreich, Sascha Wirges, Christoph Stiller
Recently, PointPainting has been presented to eliminate this performance drop by effectively fusing the output of a semantic segmentation network instead of the raw image information.
no code implementations • 26 Aug 2020 • Javier Lorenzo, Ignacio Parra, Florian Wirth, Christoph Stiller, David Fernandez Llorca, Miguel Angel Sotelo
Pedestrian crossing prediction is a crucial task for autonomous driving.
no code implementations • 13 May 2020 • Frank Bieder, Sascha Wirges, Johannes Janosovits, Sven Richter, Zheyuan Wang, Christoph Stiller
This representation allows us to use well-studied deep learning architectures from the image domain to predict a dense semantic grid map using only the sparse input data of a single LiDAR scan.
no code implementations • 9 Apr 2020 • Danial Kamran, Carlos Fernandez Lopez, Martin Lauer, Christoph Stiller
Reinforcement learning is nowadays a popular framework for solving different decision making problems in automated driving.
no code implementations • 2 Mar 2020 • Sascha Wirges, Ye Yang, Sven Richter, Haohao Hu, Christoph Stiller
We propose an object detector for top-view grid maps which is additionally trained to generate an enriched version of its input.
no code implementations • 3 Feb 2020 • Sascha Wirges, Shuxiao Ding, Christoph Stiller
We present our approach to unsupervised domain adaptation for single-stage object detectors on top-view grid maps in automated driving scenarios.
no code implementations • 30 Sep 2019 • Wei Zhan, Liting Sun, Di Wang, Haojie Shi, Aubrey Clausse, Maximilian Naumann, Julius Kummerle, Hendrik Konigshof, Christoph Stiller, Arnaud de La Fortelle, Masayoshi Tomizuka
3) The driving behavior is highly interactive and complex with adversarial and cooperative motions of various traffic participants.
no code implementations • 6 Jun 2019 • Annika Meyer, Jonas Walter, Martin Lauer, Christoph Stiller
We present our results on an evaluation set of 1000 simulated intersections and achieve 99. 9% accuracy on the topology estimation that takes only 36ms, when utilizing tracked object detections.
no code implementations • 17 Apr 2019 • Sascha Wirges, Johannes Gräter, Qiuhao Zhang, Christoph Stiller
We apply our approach to optical flow estimation from camera image sequences, validate on odometry estimation and suggest a method to iteratively increase optical flow estimation accuracy using the generated motion masks.
no code implementations • 25 Mar 2019 • Haohao Hu, Marc Sons, Christoph Stiller
To bypass the flaws from direct incorporation of GNSS measurements for geo-referencing, the usage of aerial imagery seems promising.
no code implementations • 31 Jan 2019 • Sascha Wirges, Marcel Reith-Braun, Martin Lauer, Christoph Stiller
Based on the estimated pose and shape uncertainty we approximate object hulls with bounded collision probability which we find helpful for subsequent trajectory planning tasks.
no code implementations • 2 May 2018 • Sascha Wirges, Tom Fischer, Jesus Balado Frias, Christoph Stiller
A detailed environment perception is a crucial component of automated vehicles.
no code implementations • 11 Jul 2017 • Nick Schneider, Florian Piewak, Christoph Stiller, Uwe Franke
In this paper, we present RegNet, the first deep convolutional neural network (CNN) to infer a 6 degrees of freedom (DOF) extrinsic calibration between multimodal sensors, exemplified using a scanning LiDAR and a monocular camera.
no code implementations • 19 Jun 2017 • Eike Rehder, Florian Wirth, Martin Lauer, Christoph Stiller
Accurate traffic participant prediction is the prerequisite for collision avoidance of autonomous vehicles.
no code implementations • 2 Aug 2016 • Nick Schneider, Lukas Schneider, Peter Pinggera, Uwe Franke, Marc Pollefeys, Christoph Stiller
We present a novel method for accurate and efficient up- sampling of sparse depth data, guided by high-resolution imagery.