| 摘要 |
Apparatus and a method for generating a rectified image. First pixel information corresponding to a first image is received from a first imager. Second pixel information corresponding to a second image is received from a second imager. A plurality of facial feature points of a portrait in each of the first and second images are identified. A fundamental matrix is generated based on the detected facial features. An essential matrix is generated based on the fundamental matrix. Rotational and translational information corresponding to the first and second imagers are generated based on the essential matrix. The rotational and translational information are applied to at least one of the first and second images to generate at least one rectified image.
|
