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/2304.10410
1.2. 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/2303.04302
1.3. 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/2302.06643
2. 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/2303.07430
2.2. 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/CenterFusion
2.3. 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/2107.05150
2.4. 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/2209.06535
2.5. 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/2209.06535
2.6. 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/2103.07505
2.7. 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/2004.01288
2.8. 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/2010.00058
2.9. 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/2009.08428
2.10. 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/CameraRadarFusionNet
2.11. 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/2209.13512
2.12. 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/2304.11057
2.13. 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/2304.00670
2.14. 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/2107.08948
2.15. 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/9341177
2.16. 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/2210.01346
2.17. 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/2302.10511
2.18. RA-BIRANet
题目:Radar+RGB 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/2008.13642
2.19. 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/2208.03849
2.20. 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/2208.12079
2.21. 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/2211.02432
2.22. 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/2108.10637
2.23. 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/2103.07825
2.24. 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/2003.01816
2.25. 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/2010.03320
3. 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。