自动驾驶 Camera 与 Radar 融合算法与论文总结

1. Cam与Radar融合综述论文1.1. CamRadarObjDetSemSegADSurvey题目Radar-Camera Fusion for Object Detection and Semantic Segmentation in Autonomous Driving: A Comprehensive Review名称用于自动驾驶中目标检测和语义分割的雷达相机融合综合回顾论文https://arxiv.org/abs/2304.104101.2. CamRadarPepADSurvey题目Camera-Radar Perception for Autonomous Vehicles and ADAS: Concepts, Datasets and Metrics名称自动驾驶汽车和 ADAS 的摄像头雷达感知概念、数据集和指标论文https://arxiv.org/abs/2303.043021.3. VisionRadarFusionBEVDetSurvey题目Vision-RADAR fusion for Robotics BEV Detections: A Survey名称用于机器人 BEV 检测的视觉-雷达融合一项调查论文https://arxiv.org/abs/2302.066432. Cam与Radar融合开源算法2.1. CamRadarSP题目A Modular Platform For Collaborative, Distributed Sensor Fusion名称用于协作、分布式传感器融合的模块化平台论文https://arxiv.org/abs/2303.074302.2. CenterFusion题目CenterFusion: Center-based Radar and Camera Fusion for 3D Object Detection名称CenterFusion用于 3D 对象检测的基于中心的雷达和相机融合论文https://arxiv.org/abs/2011.04841代码https://github.com/mrnabati/CenterFusion2.3. CFTrack题目CFTrack: Center-based Radar and Camera Fusion for 3D Multi-Object Tracking名称CFTrack用于 3D 多目标跟踪的基于中心的雷达和相机融合论文https://arxiv.org/abs/2107.051502.4. CRAFT题目CRAFT: Camera-Radar 3D Object Detection with Spatio-Contextual Fusion Transformer名称CRAFT使用 Spatio-Contextual Fusion Transformer 的相机-雷达 3D 目标检测论文https://arxiv.org/abs/2209.065352.5. CramNet题目CRAFT: Camera-Radar 3D Object Detection with Spatio-Contextual Fusion Transformer名称CRAFT使用 Spatio-Contextual Fusion Transformer 的相机-雷达 3D 目标检测论文https://arxiv.org/abs/2209.065352.6. CRExtCalib题目A Continuous-Time Approach for 3D Radar-to-Camera Extrinsic Calibration名称一种用于 3D 雷达到相机外部校准的连续时间方法论文https://arxiv.org/abs/2103.075052.7. CRFDriveTrj题目Extraction and Assessment of Naturalistic Human Driving Trajectories from Infrastructure Camera and Radar Sensors名称从基础设施摄像机和雷达传感器中提取和评估自然人类驾驶轨迹论文https://arxiv.org/abs/2004.012882.8. CRF-DS题目Depth Estimation from Monocular Images and Sparse Radar Data名称基于单目图像和稀疏雷达数据的深度估计论文https://arxiv.org/abs/2010.000582.9. CRF-ODDS题目Radar-Camera Sensor Fusion for Joint Object Detection and Distance Estimation in Autonomous Vehicles名称用于自动驾驶汽车联合目标检测和距离估计的雷达-相机传感器融合论文https://arxiv.org/abs/2009.084282.10. CRFNet题目A Deep Learning-based Radar and Camera Sensor Fusion Architecture for Object Detection名称用于目标检测的基于深度学习的雷达和摄像头传感器融合架构论文https://arxiv.org/abs/2005.07431代码https://github.com/TUMFTM/CameraRadarFusionNet2.11. CRF-OT题目Fusion of Inverse Synthetic Aperture Radar and Camera Images for Automotive Target Tracking名称用于汽车目标跟踪的逆合成孔径雷达和相机图像的融合论文https://arxiv.org/abs/2209.135122.12. CRF-VSM题目Vital Sign Monitoring in Dynamic Environment via mmWave Radar and Camera Fusion名称通过毫米波雷达和摄像头融合在动态环境中监测生命体征论文https://arxiv.org/abs/2304.110572.13. CRN-BEV题目CRN: Camera Radar Net for Accurate, Robust, Efficient 3D Perception名称CRN用于准确、稳健、高效 3D 感知的相机雷达网论文https://arxiv.org/abs/2304.006702.14. GenRadar题目GenRadar: Self-supervised Probabilistic Camera Synthesis based on Radar Frequencies名称GenRadar基于雷达频率的自监督概率相机合成论文https://arxiv.org/abs/2107.089482.15. GRIFNet题目GRIF Net: Gated Region of Interest Fusion Network for Robust 3D Object Detection from Radar Point Cloud and Monocular Image名称GRIF Net用于从雷达点云和单目图像进行稳健的 3D 目标检测的门控感兴趣区域融合网络论文https://ieeexplore.ieee.org/document/93411772.16. ImmFusion题目ImmFusion: Robust mmWave-RGB Fusion for 3D Human Body Reconstruction in All Weather Conditions名称ImmFusion用于全天候条件下 3D 人体重建的稳健毫米波-RGB 融合论文https://arxiv.org/abs/2210.013462.17. MVFusion题目MVFusion: Multi-View 3D Object Detection with Semantic-aligned Radar and Camera Fusion名称MVFusion使用语义对齐雷达和相机融合的多视图 3D 对象检测论文https://arxiv.org/abs/2302.105112.18. RA-BIRANet题目RadarRGB Attentive Fusion for Robust Object Detection in Autonomous Vehicles名称雷达RGB 注意力融合用于自动驾驶汽车中的鲁棒目标检测论文https://arxiv.org/abs/2008.136422.19. RadSegNet题目RadSegNet: A Reliable Approach to Radar Camera Fusion名称RadSegNet雷达相机融合的可靠方法论文https://arxiv.org/abs/2208.038492.20. RC-BEV题目Bridging the View Disparity Between Radar and Camera Features for Multi-modal Fusion 3D Object Detection名称弥合雷达和相机功能之间的视图差异用于多模态融合 3D 目标检测论文https://arxiv.org/abs/2208.120792.21. RCDPT题目RCDPT: Radar-Camera fusion Dense Prediction Transformer名称RCDPT雷达-相机融合密集预测变压器论文https://arxiv.org/abs/2211.024322.22. RCF-FVE题目Full-Velocity Radar Returns by Radar-Camera Fusion名称雷达-相机融合的全速雷达回波论文https://arxiv.org/abs/2108.106372.23. RCFusionRL题目Radar Camera Fusion via Representation Learning in Autonomous Driving名称通过自动驾驶中的表示学习融合雷达相机论文https://arxiv.org/abs/2103.078252.24. RODNet题目RODNet: Radar Object Detection Using Cross-Modal Supervision名称RODNet使用跨模态监督的雷达目标检测论文https://arxiv.org/abs/2003.018162.25. YODar题目YOdar: Uncertainty-based Sensor Fusion for Vehicle Detection with Camera and Radar Sensors名称YOdar基于不确定性的传感器融合用于使用摄像头和雷达传感器进行车辆检测论文https://arxiv.org/abs/2010.033203. 总结先前的CamRadar后/目标融合策略无法满足高阶/L3自动驾驶对功能、性能、实时、安全、鲁棒的要求。成熟的、鲁棒、高性能、高精度的基于时序的、基于BEV/Transformer/Occupancy的CamRadar前融合方案会是低成本、高阶ADAS产品落地的关键。