Dense depth information extraction method using multiple camera input images and intermediate image synthesis method

摘要

The present invention provides a method for densely extracting depth information of each pixel from images obtained from finite cameras at various positions, and efficiently using the intermediate images.;Conventional multiple-baseline stereo matching has higher accuracy than other schemes, but the computation time increases in proportion to the number of stereo pairs and the depth map obtained by fixing the size of the matching window. The boundary overreach problem of O (S) occurs, resulting in deterioration of the synthesized image quality.;The multi-resolution multi-line stereo matching method according to the present invention implements the multi-line stereo matching method in a multi-resolution method to improve the boundary time extension problem in the processing time and the depth map, and to use the neighbor candidate selection method in the multi-resolution method. By reducing the matching error for the discontinuities and providing an adaptive occlusion area processing method, the matching error is not reduced even under any camera placement situation. By reducing, depth information on each pixel of the reference image can be precisely obtained.