PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
A self-calibration method for camera using two views of unknown-structure planar scene is introduced. The planar scene is common in the environment and can be easily identifiable outside the lab. Firstly, two orientation- and scale-covariant features, which can be provided by the SIFT feature detector, is used to estimate the homography of two views. Then the homography is decomposed into the camera parameters. A RANSAC scheme is adapted to cope with the outliers of SIFT correspondences. Finally, the camera parameters are optimized with a non-linear parameter optimization using the inliers of two views. This method calibrates the camera parameters and recovers the planar scenes simultaneously. Real scene data experiment demonstrates that the proposed method is easy to operate and provides the reliable calibration results for non-expert users.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
The alert did not successfully save. Please try again later.
Qingkai Hou, Hu Chen, Zhiyun Zhuang, Fuyin Wang, Yanxin Ma, Qiong Yao, Shuidong Xiong, "Self-calibration method using the orientation- and scale-covariant features in planar scene," Proc. SPIE 11552, Optical Metrology and Inspection for Industrial Applications VII, 115521X (10 October 2020); https://doi.org/10.1117/12.2575748