I am an Associate Professor at the Intelligent Vehicles group, part of the Cognitive Robotics department of the Mechanical Engineering faculty, TU Delft, The Netherlands. You can find my staff page on the TU Delft website.

I work on multi-modal environment perception for intelligent vehicles, and mostly target crowded urban settings. My research applies deep learning and probabilistic models to automatically detect objects and predict traffic situations from multi-modal data, collected with on-vehicle sensors. Most of my research therefore relates to Computer Vision, but also addresses other sensing modalities, such as Lidar, Radar and Acoustics. To understand the environment, the autonomous vehicle needs to create useful representations of its environment, detect and identify dynamic and static obstacles, self-localize itself, and using all sensor information to anticipate future events. Applications include anticipating Vulnerable Road User (VRU) behaviour, path prediction, self-localizing the vehicle using satellite images, and fusing 3D depth and 2D image data for improved object detection.

Publications

You can find an up-to-date list on my Google Scholar profile.
Toggle screenshots

2024

  • A Deep Automotive Radar Detector using the RaDelft Dataset, I. Roldan, A. Palffy, J.F.P. Kooij, D.M. Gavrila, F. Fioranelli, A. Yaravoy, IEEE Transactions on Radar Systems, 2024. to appear
    More info …   arXiv  

  • Adapting Fine-grained Cross-view Localization to Areas without Fine Ground Truth, Z. Xia, Y. Shi, H. Li, J.F.P. Kooij, European Conference on Computer Vision (ECCV), 2024.
    More info …   arXiv   Code  
    Adapting Fine-grained Cross-view Localization to Areas without Fine Ground Truth

  • On the Estimation of Image-matching Uncertainty in Visual Place Recognition, M. Zaffar, L. Nan, J.F.P. Kooij, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 17743-17753, 2024. *Poster Highlight* (top 15% of accepted papers)
    More info …   Paper   arXiv   Code  
    On the Estimation of Image-matching Uncertainty in Visual Place Recognition

  • Convolutional Cross-View Pose Estimation, Z. Xia, O. Booij, J.F.P. Kooij, IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI), 46(5), pp. 3813-3831, 2024. doi: 10.1109/TPAMI.2023.3346924
    More info …   Paper   arXiv   Code   Video  
    Convolutional Cross-View Pose Estimation

  • UniBEV: Multi-modal 3D Object Detection with Uniform BEV Encoders for Robustness against Missing Sensor Modalities, S. Wang, H. Caesar, L. Nan, J.F.P. Kooij, IEEE Intelligent Vehicles Symposium (IVS), 2024.
    More info …   arXiv  
    UniBEV: Multi-modal 3D Object Detection with Uniform BEV Encoders for Robustness against Missing Sensor Modalities

  • Multi-Class Trajectory Prediction in Urban Traffic Using the View-of-Delft Prediction Dataset, H. Boekema, B. Martens, J.F.P. Kooij, D.M. Gavrila, IEEE Robotics and Automation Letters (RA-L), 9(5), pp. 4806-4813, 2024. doi: 10.1109/LRA.2024.3385693
    More info …   Paper   Project page  
    Multi-Class Trajectory Prediction in Urban Traffic Using the View-of-Delft Prediction Dataset

  • Adaptive Cruise Control Utilizing Noisy Multi-leader Measurements: A Learning-based Approach, Y. Ni, V.L. Knoop, J.F.P. Kooij, B. van Arem, IEEE Open Journal of Intelligent Transportation Systems, 5, pp. 251-264, 2024. doi: 10.1109/OJITS.2024.3395149
    More info …   Paper  
    Adaptive Cruise Control Utilizing Noisy Multi-leader Measurements: A Learning-based Approach

  • See further than CFAR: A data-driven radar detector trained by Lidar, I. Roldan, A. Palffy, J.F.P. Kooij, D.M. Gavrila, F. Fioranelli, A. Yaravoy, IEEE Radar Conference, 2024. *received Best Student Paper award*
    More info …   arXiv  
    See further than CFAR: A data-driven radar detector trained by Lidar

  • MuVieCAST: Multi-View Consistent Artistic Style Transfer, N. Ibrahimli, J.F.P. Kooij, L. Nan, IEEE International Conference on 3D Vision (3DV), 2024. to appear
    More info …   arXiv   Code   Project page  
    MuVieCAST: Multi-View Consistent Artistic Style Transfer

2023

  • DDL-MVS: Depth Discontinuity Learning for Multi-View Stereo Networks, N. Ibrahimli, H. Ledoux, J.F.P. Kooij, L. Nan, Remote Sensing, 15, pp. 2970, 2023. doi: 10.3390/rs15122970
    More info …   Paper   arXiv   Code  
    DDL-MVS: Depth Discontinuity Learning for Multi-View Stereo Networks

  • SliceMatch: Geometry-guided Aggregation for Cross-View Pose Estimation, T. Lentsch, Z. Xia, H. Caesar, J.F.P. Kooij, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 17225-17234, 2023.
    More info …   Paper   arXiv   Code   Video  
    SliceMatch: Geometry-guided Aggregation for Cross-View Pose Estimation

  • CoPR: Toward Accurate Visual Localization With Continuous Place-descriptor Regression, M. Zaffar, L. Nan, J.F.P. Kooij, IEEE Transactions on Robotics (T-RO), 39(4), pp. 2825-2841, 2023. doi: 10.1109/TRO.2023.3262106
    More info …   Paper   arXiv  
    CoPR: Toward Accurate Visual Localization With Continuous Place-descriptor Regression

  • How Informative is the Approximation Error from Tensor Decomposition for Neural Network Compression?, J. Schuurmans, K. Batselier, J.F.P. Kooij, International Conference on Learning Representations (ICLR), 2023. doi: 10.48550/arXiv.2305.05318
    More info …   Paper   arXiv   Code  
    How Informative is the Approximation Error from Tensor Decomposition for Neural Network Compression?

  • Detecting darting out pedestrians with occlusion aware sensor fusion of radar and stereo camera, A. Palffy, J.F.P. Kooij, D.M. Gavrila, IEEE Transactions on Intelligent Vehicles (T-IV), 8(2), pp. 1459-1472, 2023. doi: 10.1109/TIV.2022.3220435
    More info …   Paper  
    Detecting darting out pedestrians with occlusion aware sensor fusion of radar and stereo camera

2022

  • Push-the-Boundary: Boundary-aware Feature Propagation for Semantic Segmentation of 3D Point Clouds, S. Du, N. Ibrahimli, J. Stoter, J. Kooij, L. Nan, IEEE International Conference on 3D Vision (3DV), 2022.
    More info …   arXiv   Code   Video  
    Push-the-Boundary: Boundary-aware Feature Propagation for Semantic Segmentation of 3D Point Clouds

  • Visual Cross-View Metric Localization with Dense Uncertainty Estimates, Z. Xia, O. Booij, M. Manfredi, J.F.P. Kooij, European Conference on Computer Vision (ECCV), pp. 90-106, 2022. doi: 10.1007/978-3-031-19842-7_6
    More info …   Paper   arXiv   Code  
    Visual Cross-View Metric Localization with Dense Uncertainty Estimates

  • Multi-class Road User Detection with 3+1D Radar in the View-of-Delft Dataset, A. Palffy, E.A.I. Pool, S. Baratam, J.F.P. Kooij, D.M. Gavrila, IEEE Robotics and Automation Letters (RA-L), 7(2), pp. 4961-4968, 2022. doi: 10.1109/LRA.2022.3147324 presented at IEEE ICRA 2022
    More info …   Paper   Video   Project page  
    Multi-class Road User Detection with 3+1D Radar in the View-of-Delft Dataset

  • BackboneAnalysis: Structured Insights into Compute Platforms from CNN Inference Latency, F.M. Hafner, M. Zeller, M. Schutera, J. Abhau, J.F.P. Kooij, IEEE Intelligent Vehicles Symposium (IVS), pp. 1801-1809, 2022. doi: 10.1109/IV51971.2022.9827260
    More info …   Paper  
    BackboneAnalysis: Structured Insights into Compute Platforms from CNN Inference Latency

  • Cross-modal distillation for RGB-depth person re-identification, F. M. Hafner, A. Bhuyian, J.F.P. Kooij, E. Granger, Computer Vision and Image Understanding (CVIU), 216, pp. 103352, 2022. doi: 10.1016/j.cviu.2021.103352
    More info …   Paper   arXiv  
    Cross-modal distillation for RGB-depth person re-identification

2021

  • Occupancy Grid Mapping with Cognitive Plausibility for Autonomous Driving Applications, A. Plebe, J.F.P. Kooij, G.P.R. Papini, M. Da Lio, Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, pp. 2934-2941, 2021.
    More info …   Paper  
    Occupancy Grid Mapping with Cognitive Plausibility for Autonomous Driving Applications

  • Cross-View Matching for Vehicle Localization by Learning Geographically Local Representations, Z. Xia, O. Booij, M. Manfredi, J.F.P. Kooij, IEEE Robotics and Automation Letters (RA-L), 6(3), pp. 5921-5928, 2021. doi: 10.1109/LRA.2021.3088076
    More info …   Paper   Code   Video  
    Cross-View Matching for Vehicle Localization by Learning Geographically Local Representations

  • VPR-Bench: An Open-Source Visual Place Recognition Evaluation Framework with Quantifiable Viewpoint and Appearance Change, M. Zaffar, S. Garg, M. Milford, J. Kooij, D. Flynn, K. McDonald-Maier, S. Ehsan, International Journal of Computer Vision (IJCV), 2021. doi: 10.1007/s11263-021-01469-5
    More info …   Paper   arXiv   Code  
    VPR-Bench: An Open-Source Visual Place Recognition Evaluation Framework with Quantifiable Viewpoint and Appearance Change

  • Fast and Compact Image Segmentation using Instance Stixels, T.H. Hehn, J.F.P. Kooij, D.M. Gavrila, IEEE Transactions on Intelligent Vehicles (T-IV), 7(1), pp. 45-46, 2021. doi: 10.1109/TIV.2021.3067223
    More info …   Paper   Code   Video  
    Fast and Compact Image Segmentation using Instance Stixels

  • Crafted vs. Learned Representations in Predictive Models - A Case Study on Cyclist Path Prediction, E.A.I. Pool, J.F.P.Kooij, D.M. Gavrila, IEEE Transactions on Intelligent Vehicles (T-IV), 6(4), pp. 747-759, 2021. doi: 10.1109/TIV.2021.3064253
    More info …   Paper  
    Crafted vs. Learned Representations in Predictive Models - A Case Study on Cyclist Path Prediction

  • Hearing What You Cannot See: Acoustic Vehicle Detection Around Corners, Y. Schulz, A.K. Mattar, T.M. Hehn, J.F.P. Kooij, IEEE Robotics and Automation Letters (RA-L), 6(2), pp. 2587-2594, 2021. doi: 10.1109/LRA.2021.3062254 presented at IEEE ICRA 2021
    More info …   Paper   arXiv   Code   Video  
    Hearing What You Cannot See: Acoustic Vehicle Detection Around Corners

  • A Joint Extrinsic Calibration Tool for Radar, Camera and Lidar, J. Domhof, J. Kooij, D. Gavrila, IEEE Transactions on Intelligent Vehicles (T-IV), 6(3), pp. 571-582, 2021. doi: 10.1109/TIV.2021.3065208
    More info …   Paper   Code  
    A Joint Extrinsic Calibration Tool for Radar, Camera and Lidar

2020

2019

  • RGB-Depth Cross-Modal Person Re-identification, F.M. Hafner, A. Bhuiyan, J.F.P. Kooij, E. Granger, IEEE International Conference on Advanced Video and Signal-based Surveillance (AVSS), pp. 1-8, 2019. doi: 10.1109/AVSS.2019.8909838 oral
    More info …   Paper  
    RGB-Depth Cross-Modal Person Re-identification

  • Context-Based Cyclist Path Prediction Using Recurrent Neural Networks, E.A.I. Pool, J.F.P. Kooij, D.M. Gavrila, IEEE Intelligent Vehicles Symposium (IVS), pp. 824-830, 2019. doi: 10.1109/IVS.2019.8813889 oral
    More info …   Paper  
    Context-Based Cyclist Path Prediction Using Recurrent Neural Networks

  • SafeVRU: A Research Platform for the Interaction of Self-Driving Vehicles with Vulnerable Road Users, L. Ferranti, B. Brito, E.A.I. Pool, Y. Zhang, R.M. Ensing, R. Happee, B. Shyrokau, J.F.P. Kooij, J. Alonso-Mora, D.M. Gavrila, IEEE Intelligent Vehicles Symposium (IVS), pp. 1660-1666, 2019. doi: 10.1109/IVS.2019.8813899
    More info …   Paper   Video  
    SafeVRU: A Research Platform for the Interaction of Self-Driving Vehicles with Vulnerable Road Users

  • Instance Stixels: Segmenting and Grouping Stixels into Objects, T.H. Hehn, J.F.P. Kooij, D.M. Gavrila, IEEE Intelligent Vehicles Symposium (IVS), pp. 2542-2549, 2019. doi: 10.1109/IVS.2019.8814243
    More info …   Paper   Code   Video  
    Instance Stixels: Segmenting and Grouping Stixels into Objects

  • Occlusion Aware Sensor Fusion for Early Crossing Pedestrian Detection, A. Palffy, J.F.P. Kooij, D.M. Gavrila, IEEE Intelligent Vehicles Symposium (IVS), pp. 1768-1774, 2019. doi: 10.1109/IVS.2019.8814065
    More info …   Paper  
    Occlusion Aware Sensor Fusion for Early Crossing Pedestrian Detection

  • An Extrinsic Calibration Tool for Lidar, Camera and Radar, J. Domhof, J.F.P. Kooij, D.M. Gavrila, IEEE International Conference on Robotics and Automation (ICRA), pp. 8107-8113, 2019. doi: 10.1109/ICRA.2019.8794186
    More info …   Paper   Code  
    An Extrinsic Calibration Tool for Lidar, Camera and Radar

2018

  • Context-Based Path Prediction for Targets with Switching Dynamics, J.F.P. Kooij, F. Flohr, E.A.I. Pool, D.M. Gavrila, International Journal of Computer Vision (IJCV), 2018. doi: 10.1007/s11263-018-1104-4
    More info …   Paper  
    Context-Based Path Prediction for Targets with Switching Dynamics

2017

2016

2015

  • A probabilistic framework for joint pedestrian head and body orientation estimation, F. Flohr, M. Dumitru-Guzu, JFP Kooij, IEEE Transactions on Intelligent Transportation Systems (T-ITS), 16(4), pp. 1872-1882, 2015. doi: 10.1109/TITS.2014.2379441
    More info …   Paper  
    A probabilistic framework for joint pedestrian head and body orientation estimation

  • Identifying multiple objects from their appearance in inaccurate detections, J.F.P. Kooij, G. Englebienne, D.M. Gavrila, Computer Vision and Image Understanding (CVIU), 136, pp. 103-116, 2015. doi: 10.1016/j.cviu.2015.03.012
    More info …   Paper  
    Identifying multiple objects from their appearance in inaccurate detections

2014

2012

  • A Non-parametric Hierarchical Model to Discover Behavior Dynamics from Tracks, J.F.P. Kooij, G. Englebienne, D.M. Gavrila, Proceedings of the European Conference on Computer Vision (ECCV), pp. 270-283, 2012. doi: 10.1007/978-3-642-33783-3_20
    More info …   Paper  
    A Non-parametric Hierarchical Model to Discover Behavior Dynamics from Tracks

2008

  • The cross-entropy method for policy search in decentralized POMDPs, F.A. Oliehoek, J.F.P. Kooij, N. Vlassis, Informatica, 32(4), pp. 341-357, 2008.
    More info …   Paper  
    The cross-entropy method for policy search in decentralized POMDPs

  • A cross-entropy approach to solving Dec-POMDPs, F.A. Oliehoek, J.F.P. Kooij, N. Vlassis, Advances in Intelligent and Distributed Computing, pp. 145-154, 2008. doi: 10.1007/978-3-540-74930-1_15
    More info …   Paper  
    A cross-entropy approach to solving Dec-POMDPs