YOdar: Uncertainty-based Sensor Fusion for Vehicle Detection with Camera and Radar Sensors

In this work, we present an uncertainty-based method for sensor fusion with camera and radar data. The outputs of two neural networks, one processing camera and the other one radar data, are combined in an uncertainty aware manner. To this end, we gather the outputs and corresponding meta information for both networks. For each predicted object, the gathered information is post-processed by a gradient boosting method to produce a joint prediction of both networks. In our experiments we combine the YOLOv3 object detection network with a customized $1D$ radar segmentation network and evaluate our method on the nuScenes dataset. In particular we focus on night scenes, where the capability of object detection networks based on camera data is potentially handicapped. Our experiments show, that this approach of uncertainty aware fusion, which is also of very modular nature, significantly gains performance compared to single sensor baselines and is in range of specifically tailored deep learning based fusion approaches.

[1]  Erwin Biebl,et al.  A machine learning joint lidar and radar classification system in urban automotive scenarios , 2019 .

[2]  Ahmet M. Kondoz,et al.  Robust Fusion of LiDAR and Wide-Angle Camera Data for Autonomous Mobile Robots , 2017, Sensors.

[3]  J. Friedman Stochastic gradient boosting , 2002 .

[4]  Hanno Gottschalk,et al.  Prediction Error Meta Classification in Semantic Segmentation: Detection via Aggregated Dispersion Measures of Softmax Probabilities , 2018, 2020 International Joint Conference on Neural Networks (IJCNN).

[5]  Hui Kong,et al.  Road Detection through CRF based LiDAR-Camera Fusion , 2019, 2019 International Conference on Robotics and Automation (ICRA).

[6]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  Erwin Biebl,et al.  Deep Learning Radar Object Detection and Classification for Urban Automotive Scenarios , 2019, 2019 Kleinheubach Conference.

[8]  Ali Farhadi,et al.  YOLOv3: An Incremental Improvement , 2018, ArXiv.

[9]  Zhangjing Wang,et al.  Multi-Sensor Fusion in Automated Driving: A Survey , 2020, IEEE Access.

[10]  Pietro Perona,et al.  Microsoft COCO: Common Objects in Context , 2014, ECCV.

[11]  Hanno Gottschalk,et al.  Time-Dynamic Estimates of the Reliability of Deep Semantic Segmentation Networks , 2019, 2020 IEEE 32nd International Conference on Tools with Artificial Intelligence (ICTAI).

[12]  Ranjit Sadakale,et al.  Monocular Camera based Computer Vision System for Cost Effective Autonomous Vehicle , 2019, 2019 10th International Conference on Computing, Communication and Networking Technologies (ICCCNT).

[13]  Markus Lienkamp,et al.  A Deep Learning-based Radar and Camera Sensor Fusion Architecture for Object Detection , 2019, 2019 Sensor Data Fusion: Trends, Solutions, Applications (SDF).

[14]  Guodong Rong,et al.  LGSVL Simulator: A High Fidelity Simulator for Autonomous Driving , 2020, 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC).

[15]  Vijay John,et al.  RVNet: Deep Sensor Fusion of Monocular Camera and Radar for Image-Based Obstacle Detection in Challenging Environments , 2019, PSIVT.

[16]  Jiun-In Guo,et al.  LiDAR/camera sensor fusion technology for pedestrian detection , 2017, 2017 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC).

[17]  Matthias Rottmann,et al.  MetaDetect: Uncertainty Quantification and Prediction Quality Estimates for Object Detection , 2020, 2021 International Joint Conference on Neural Networks (IJCNN).

[18]  Anis Koubaa,et al.  Car Detection using Unmanned Aerial Vehicles: Comparison between Faster R-CNN and YOLOv3 , 2018, 2019 1st International Conference on Unmanned Vehicle Systems-Oman (UVS).

[19]  Nanning Zheng,et al.  Detecting Drivable Area for Self-driving Cars: An Unsupervised Approach , 2017, ArXiv.

[20]  Hanno Gottschalk,et al.  Controlled False Negative Reduction of Minority Classes in Semantic Segmentation , 2019, 2020 International Joint Conference on Neural Networks (IJCNN).

[21]  Matthias Rottmann,et al.  Uncertainty Measures and Prediction Quality Rating for the Semantic Segmentation of Nested Multi Resolution Street Scene Images , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[22]  Germán Ros,et al.  CARLA: An Open Urban Driving Simulator , 2017, CoRL.

[23]  Thanuka Wickramarathne,et al.  Low-Cost Radar for Object Tracking in Autonomous Driving: A Data-Fusion Approach , 2018, 2018 IEEE 87th Vehicular Technology Conference (VTC Spring).

[24]  Jihua Ye,et al.  Making of Night Vision: Object Detection Under Low-Illumination , 2020, IEEE Access.

[25]  Bernardo Wagner,et al.  Radar and LiDAR Sensorfusion in Low Visibility Environments , 2016, ICINCO.

[26]  Qiang Xu,et al.  nuScenes: A Multimodal Dataset for Autonomous Driving , 2019, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[27]  Fabian de Ponte Müller,et al.  Survey on Ranging Sensors and Cooperative Techniques for Relative Positioning of Vehicles , 2017, Sensors.

[28]  P. Tumas,et al.  Pedestrian Detection in Severe Weather Conditions , 2020, IEEE Access.

[29]  Mikkel Kragh Hansen,et al.  Multi-Modal Obstacle Detection in Unstructured Environments with Conditional Random Fields , 2020, J. Field Robotics.

[30]  Guofa Li,et al.  Deep Learning Approaches on Pedestrian Detection in Hazy Weather , 2020, IEEE Transactions on Industrial Electronics.

[31]  Martin Schneider,et al.  Automotive Radar – Status and Trends , 2005 .