METHOD FOR ESTIMATING THE LOCATION OF AN IMAGE TARGET REGION FROM TRACKED MULTIPLE IMAGE LANDMARK REGIONS

摘要

An image processing method for inserting a given pattern at a target region (304A) having a particular location with respect to a scene being viewed by an image sensor (300A) over a period of time, wherein the method employs a world map (332) having stored therein the relative position of the location and pose of multiple pre-trained reference image patterns of landmark regions (A, B, C, D, and E) in the scene with respect to that of the target region. The method comprises dynamic computation steps for inferring the size and position of the particular location within each ongoing successive image frames of the scene from the shape, size and position of at least one of said multiple landmark regions represented within each of successive image frames of the scene, despite inaccuracies in the parametric model estimation relating the current image with pre-trained reference image and changes over time in the shape, size and position.