You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, first of all thank you very much for the tool and the in depth documentation / calibration troubleshooting guide.
Performing the calibration dance with two big baseline cameras can be challenging because all chessboard corners should be detected. We are using Charuco boards.
Although the documentation states clearly that all chessboard corners should be detected, the calibration.py comments mentions that one could give weight 0 in order to support incomplete detections. Would this work?
The text was updated successfully, but these errors were encountered:
The mrgingham detector requires the whole chessboard to be visible in order for the detection to succeed. But mrcal does not require this detector to be used. You can utilize any other detector, and feed the (possibly-incomplete) observations to mrcal. See https://mrcal.secretsauce.net/recipes.html#non-mrgingham-detector
Let me know if you have questions. If not, please close the report. Thanks
Hi, first of all thank you very much for the tool and the in depth documentation / calibration troubleshooting guide.
Performing the calibration dance with two big baseline cameras can be challenging because all chessboard corners should be detected. We are using Charuco boards.
Although the documentation states clearly that all chessboard corners should be detected, the calibration.py comments mentions that one could give weight 0 in order to support incomplete detections. Would this work?
The text was updated successfully, but these errors were encountered: