ethz-asl / kalibr

The Kalibr visual-inertial calibration toolbox
Other
4.28k stars 1.39k forks source link

The cameras are not connected through mutual target observations! Please provide another dataset... #554

Closed ghost closed 2 years ago

ghost commented 2 years ago

标定D435i的时候报错,无法生成标定数据,需要提供另一个dataset,这是我的报错信息:


rosrun kalibr kalibr_calibrate_cameras --target checkerboard.yaml --bag mult_cam_d435i.bag --models pinhole-radtan pinhole-radtan pinhole-radtan --topics /color /infra_left /infra_right importing libraries Initializing cam0: Camera model: pinhole-radtan Dataset: mult_cam_d435i.bag Topic: /color Number of images: 543 Extracting calibration target corners Extracted corners for 543 images (of 543 images)
Projection initialized to: [ 895.28080899 895.40661113 647.75742412 359.02253582] Distortion initialized to: [ 0.11513699 -0.23354251 0.00048682 -0.0005654 ] Initializing cam1: Camera model: pinhole-radtan Dataset: mult_cam_d435i.bag Topic: /infra_left Number of images: 543 Extracting calibration target corners Extracted corners for 543 images (of 543 images)
Projection initialized to: [ 417.3526174 418.87301196 427.81256676 243.43318206] Distortion initialized to: [-0.00517335 -0.00251272 0.0013305 0.00090247] Initializing cam2: Camera model: pinhole-radtan Dataset: mult_cam_d435i.bag Topic: /infra_right Number of images: 543 Extracting calibration target corners Extracted corners for 543 images (of 543 images)
Projection initialized to: [ 417.38083763 419.13104002 427.76249859 245.09232602] Distortion initialized to: [ 0.00173417 -0.00973484 0.00148911 0.00111103] initializing initial guesses [ERROR] [1659680833.373050]: The cameras are not connected through mutual target observations! Please provide another dataset...


goldbattle commented 2 years ago

Please provide more information and a bag file. Likely the issue is what the error reports. "The cameras are not connected through mutual target observations!" This means the same board was not seen from both cameras at the same time. This could be either your cameras don't have a FOV overlap or their timestamps are incorrect.

Additionally, please speak english here.