3D image generation

摘要

A system generating a 3D depth profile of an object, including: a moveable video camera to continually capture 2D image data of the object as the camera moves, to capture plural 2D image frames each including plural pixels; a processor configured to: receive the 2D image data; determine position of the camera when each frame is captured; calculate depth of part of an object shown in a pixel in a first frame with respect to a reference, by identifying the part in at least one further frame and calculating the depth using camera positions of the first image and at least one further image; and determine depth profile of the object, for plural parts of the object shown in the pixels. A display displays the depth profile and indicates in real time, as data is gathered, the parts of the object for which the depth profile has been calculated.

主权项

1. A system for generating a 3D depth profile of an object, the system comprising:
a moveable video camera configured to continually capture 2D image data of said object as said camera is moved, said image data being captured as a plurality of 2D image frames where each frame comprises a plurality of pixels; a processor configured: to receive said 2D image data; to determine the position of the camera for each frame captured; to calculate the depth of the part of an object shown in a pixel in a first frame, said depth being determined with respect to a reference, said depth being determined by identifying said part of the object in at least one further frame and calculating the depth using the camera positions of said first frame and said at least one further frame,
wherein said depth is calculated by determining the similarity of a pixel in said first frame with a plurality of pixels in said at least one further frame, said plurality of pixels in said further frame being located along an epipolar line in said further frame, said epipolar line representing the possible depths of said part of the object;wherein a matching score indicating the similarity of a pixel in the first frame with a pixel in said further frame is determined for the pixels along said epipolar line, and potential matches for said pixel in the first frame are selected as local maxima for said matching score when dependent on depth,wherein said matching scores as a function of depth for a plurality of images are combined to determine a single value for the depth of said part of object,wherein the distribution of the matching score against depth is modelled as a mixture of a uniform distribution and a unimodal distribution,wherein a weighting is given to said unimodal distribution, said weighting indicating the probability that the depth indicated by said unimodal distribution is a good measurement;wherein the parameters of said model are estimated using Bayesian Posterior Estimation;wherein said model is approximated by a distribution of the type:
Cxp(xt|Z,π)p(Z,π|at−1,bt−1,μt−1,σt−12)where xt is a tth point, Z is the depth, π is the weighting, a and b are parameters of a beta distribution and μ and σ2 are the mean and standard deviation of said unimodal distribution; and
to determine a depth profile of said object, the depth profile being determined for a plurality of parts of the object shown in said pixels, the system further comprises a display configured to display said depth profile and an indication in real time as data is still being gathered, the parts of the object for which the depth profile has been calculated.