Litcius/Paper detail

CalibRCNN: Calibrating Camera and LiDAR by Recurrent Convolutional Neural Network and Geometric Constraints

Jieying Shi, Ziheng Zhu, Jianhua Zhang, Ruyu Liu, Zhenhua Wang, Shengyong Chen, Honghai Liu

202073 citationsDOIOpen Access PDF

Abstract

In this paper, we present Calibration Recurrent Convolutional Neural Network (CalibRCNN) to infer a 6 degrees of freedom (DOF) rigid body transformation between 3D LiDAR and 2D camera. Different from the existing methods, our 3D-2D CalibRCNN not only uses the LSTM network to extract the temporal features between 3D point clouds and RGB images of consecutive frames, but also uses the geometric loss and photometric loss obtained by the interframe constraint to refine the calibration accuracy of the predicted transformation parameters. The CalibRCNN aims at inferring the correspondence between projected depth image and RGB image to learn the underlying geometry of 2D-3D calibration. Thus, the proposed calibration model achieves a good generalization ability to adapt to unknown initial calibration error ranges, and other 3D LiDAR and 2D camera pairs with different intrinsic parameters from the training dataset. Extensive experiments have demonstrated that our CalibRCNN can achieve state-of-the-art accuracy by comparison with other CNN based methods.

Topics & Concepts

Artificial intelligenceConvolutional neural networkComputer scienceComputer visionCalibrationLidarRGB color modelPoint cloudTransformation (genetics)Camera resectioningGeneralizationConstraint (computer-aided design)Pattern recognition (psychology)Remote sensingMathematicsGeometryBiochemistryMathematical analysisChemistryGeneStatisticsGeologyAdvanced Vision and ImagingOptical measurement and interference techniquesRobotics and Sensor-Based Localization
CalibRCNN: Calibrating Camera and LiDAR by Recurrent Convolutional Neural Network and Geometric Constraints | Litcius