PJLab-ADG / SensorsCalibration

OpenCalib: A Multi-sensor Calibration Toolbox for Autonomous Driving
Apache License 2.0
2.29k stars 545 forks source link

关于Lidar2Cam中autocalib在kitti上提取不到特征的问题 #80

Open Chenwei-Huang opened 1 year ago

Chenwei-Huang commented 1 year ago

您好,我运行了你们的testdata,效果非常不错 但是我按照论文换用了kitti数据集((09/26/2011 drive 0015),结果lidar和image都检测不到特征了,请问是需要改什么设置吗 下面是运行时的打印日志: center_camera-intrinsic top_center_lidar-to-center_camera-extrinsic [pcl::SampleConsensusModel::getSamples] Can not select 0 unique points out of 0! [pcl::RandomSampleConsensus::computeModel] No samples could be selected! [pcl::SACSegmentation::segment] Error segmenting the model! No solution found. [WARN] [122@lidar_feature.cpp] Could not estimate a planar model for the given dataset. before loss: -nan after loss: -nan

同样的,python运行的图像特征提取的mask也是全黑图片,并没有车道线和电线杆 测试了多组kitti数据都是这样,所以想问一下,如果想要达到论文中的kitti数据集标定效果,有什么需要注意的地方吗? 感谢!!

xiaokn commented 1 year ago

您好,我运行了你们的testdata,效果非常不错 但是我按照论文换用了kitti数据集((09/26/2011 drive 0015),结果lidar和image都检测不到特征了,请问是需要改什么设置吗 下面是运行时的打印日志: center_camera-intrinsic top_center_lidar-to-center_camera-extrinsic [pcl::SampleConsensusModel::getSamples] Can not select 0 unique points out of 0! [pcl::RandomSampleConsensus::computeModel] No samples could be selected! [pcl::SACSegmentation::segment] Error segmenting the model! No solution found. [WARN] [122@lidar_feature.cpp] Could not estimate a planar model for the given dataset. before loss: -nan after loss: -nan

同样的,python运行的图像特征提取的mask也是全黑图片,并没有车道线和电线杆 测试了多组kitti数据都是这样,所以想问一下,如果想要达到论文中的kitti数据集标定效果,有什么需要注意的地方吗? 感谢!!

是不是配置的问题?

cchester25 commented 1 year ago

我也遇到了这个问题

您好,我运行了你们的testdata,效果非常不错 但是我按照论文换用了kitti数据集((09/26/2011 drive 0015),结果lidar和image都检测不到特征了,请问是需要改什么设置吗 下面是运行时的打印日志: center_camera-intrinsic top_center_lidar-to-center_camera-extrinsic [pcl::SampleConsensusModel::getSamples] Can not select 0 unique points out of 0! [pcl::RandomSampleConsensus::computeModel] No samples could be selected! [pcl::SACSegmentation::segment] Error segmenting the model! No solution found. [WARN] [122@lidar_feature.cpp] Could not estimate a planar model for the given dataset. before loss: -nan after loss: -nan 同样的,python运行的图像特征提取的mask也是全黑图片,并没有车道线和电线杆 测试了多组kitti数据都是这样,所以想问一下,如果想要达到论文中的kitti数据集标定效果,有什么需要注意的地方吗? 感谢!!

是不是配置的问题?

我也遇到了这个问题,难道不同的照片都需要再重新训练一遍么?