Automated three dimensional mapping method

摘要

An automated three dimensional mapping method estimating three dimensional models taking advantage of a plurality of images. Positions and attitudes for at least one camera are recorded when images are taken. The at least one camera is geometrically calibrated to indicate the direction of each pixel of an image. A stereo disparity is calculated for a plurality of image pairs covering a same scene position setting a disparity and a certainty measure estimate for each stereo disparity. The different stereo disparity estimates are weighted together to form a 3D model. The stereo disparity estimates are reweighted automatically and adaptively based on the estimated 3D model.

主权项

1. An automated three dimensional mapping method estimating three dimensional models covering the whole area covered by collected images without selection of key points or segmentation of objects and taking advantage of a plurality of images, the method comprising:
recording positions and attitudes for at least one camera when images are taken along a path; geometrically calibrating at least one camera to indicate the direction from each pixel in an image; calculating a pixel wise stereo disparity for a plurality of image pairs covering a same scene position setting a disparity and a certainty measure estimate for each stereo disparity; weighing together the different stereo disparity estimates based on the certainty measures to form a 3D model, wherein the certainty measure estimate for each disparity is set in consideration of resolution, local contrast, or visibility; and automatically and adaptively reweighing the stereo disparity estimates based on the estimated 3D model and the certainty measures, wherein said reweighing is based on the normal vectors of the estimated 3D model.