RGB Image‑ and Lidar‑Based 3D Object Detection Under Multiple Lighting Scenarios

作     者：Wentao Chen Wei Tian Xiang Xie Wilhelm Stork 

作者机构：School of Automotive StudiesTongji UniversityShanghaiChina Karlsruhe Institute of Technology76131 KarlsruheGermany 

出 版 物：《Automotive Innovation》 (汽车创新工程（英文）)

年 卷 期：2022年第5卷第3期

页      面：251-259页

核心收录：

学科分类：0810[工学-信息与通信工程] 08[工学] 080203[工学-机械设计及理论] 0804[工学-仪器科学与技术] 0802[工学-机械工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by the National Natural Science Foundation of China(No.52002285) the Shanghai Pujiang Program(No.2020PJD075) the Natural Science Foundation of Shanghai(No.21ZR1467400) 

主　　题：3D object detection Multi-sensor fusion Uncertainty estimation Semantic segmentation PointPainting 

摘      要：In recent years,camera-and lidar-based 3D object detection has achieved great ***,the related researches mainly focus on normal illumination conditions;the performance of their 3D detection algorithms will decrease under low lighting scenarios such as in the *** work attempts to improve the fusion strategies on 3D vehicle detection accuracy in multiple lighting ***,distance and uncertainty information is incorporated to guide the“paintingof semantic information onto point cloud during the data ***,a multitask framework is designed,which incorpo-rates uncertainty learning to improve detection accuracy under low-illumination *** the validation on KITTI and Dark-KITTI benchmark,the proposed method increases the vehicle detection accuracy on the KITTI benchmark by 1.35%and the generality of the model is validated on the proposed Dark-KITTI dataset,with a gain of 0.64%for vehicle detection.