Learning-based Camera Calibration

A Learning-based Camera Calibration Toolbox

Paper

The pdf file can be found here.

@misc{zhang2022learningbased,
    title={Learning-Based Framework for Camera Calibration with Distortion Correction and High Precision Feature Detection},
    author={Yesheng Zhang and Xu Zhao and Dahong Qian},
    year={2022},
    eprint={2202.00158},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

Update log

• 22-02-01: Code is coming soon...

• 21-06-15: You can run the Demo_calib.py for calibration demo on synthetic data after modifying the path set in it (The demo data is saved in \CCS\Code\demo_data).

• 21-07-19: RANSAC-based calibration by Zhang's method is added as Calib.calib_RANSAC_OpenCV().

TODO LIST

• code release.
• README complete.

File Folder Configuration

See ./demo_data/demo_normal.

Workflow

• Distortion Correction
  If your images suffer from severe radial lens distortion, you are recommended to perfrom distortion correction as follow.
  To be continue ...

• Feature Extraction Use our network combined with surface fitting algorithm to extract chessborad features and achieve sub-pixel accuracy.