A Camera Calibration Method Based on Real-Time Video

Camera calibration is one of the most important steps in computer vision. This paper presents a camera calibration algorithm based on real-time video. Video key frames are extracted by inter-frame difference algorithm, and camera calibration is carried out by key frames extracted from video. The influence of background on key frame extraction is eliminated. Moreover, the calibration process is self-calibrated, which makes the process more convenient, fast, and practical. The uncertainty of selecting images is artificially eliminated and the results of camera calibration are improved.