Search academic texts

Information × Registration Number 2120U007872, Article popup.category Препринт Title Manhattan Frame Detection in Lens Distorted Images (AI translated) popup.author Liepieshov KostiantynLiepieshov Kostiantyn popup.publication 01-01-2020 popup.source_user Український католицький університет popup.source https://hdl.handle.net/20.500.14570/4510 popup.publisher Description Camera auto-calibration from a single image with radial distortion is a prevalent task in computer vision. Most of the existing approaches are based on the same process of extraction of features, such as circles, from the image. Since those features are noisy, the error is propagated to the higher level, and the final estimations are inaccurate. We incorporate the constraints imposed by the division model of radial distor- tion and suggest a simple approach that gives soft estimates of three Manhattan di- rections. For this task, we adapt a well-known Expectation Maximisation algorithm. We combine it with different initialization and filtering steps that we form based on the division model and Manhattan world assumptions. We illustrate the performance of the proposed approach on YORK Urban Database (YUD) and AIT Dataset of indoor and outdoor scenes. Besides, we experiment with the proposed initializations and filtering procedures. popup.nrat_date 2025-11-05 Close