Method and device for generating 3d images
Abstract
A method and a device for generating 3D images, according to which an image of a second sequence of images is generated in addition to an image of a first sequence of 2D images at an interval that can be determined via an approximation variable (α). A measure of similarity (d k ) between successive images of the first sequence is determined and compared with threshold values (δ 0 <δ 1 <δ 2 ) so as to modify the approximation variable (α) depending thereon in such a manner that the stereo base width does not turn unnaturally large. A phase analyzer ( 12 ) is used to determine a prevailing direction of movement in successive images of the first sequence of images and a phase converter ( 16 ) is used to allocate the image of the first and second sequence of images to a left-hand or right-hand viewing channel depending on a prevailing direction of movement in successive images of the first sequence.
Claims
exact text as granted — not AI-modified1 - 14 . (canceled)
15 . A method for generating 3-D images from a first sequence of 2-D images, comprising the following steps:
(a) comparing a pair of sequential images in the first image sequence to determine a measure of similarity (d k ) between the pair of sequential images; (b) comparing the measure of similarity (d k ) to predetermined threshold values (δ 0 <δ 1 <δ 2 ); (c) if δ 1 <δ k <δ 2 , (i) setting an approximation variable (a) within a predetermined range to maintain a minimum stereo base width and to prevent the stereo base from becoming too large, and (ii) generating a synthetic image for a second image sequence from the pair of images of the first image sequence by interpolating between the pair of images of the first image sequence based on the approximation variable; (d) if d k <δ 0 setting a temporally previous image of the first image sequence as an image of the second image sequence; and (e) assigning images of the first and second image sequence to a left and right viewing channel, respectively.
16 . A method as set forth in claim 15 , wherein if δ 1 <δ k <δ 2 and d k −d k−1 ≦−δ 1 and as long as α≦k−1, the approximation variable is set to α:=α+s, and if δ 1 <d k <δ 2 and d k −d k-1 ≧δ 1 and as long as α>2, the approximation variable is set to α:=α−s.
17 . A method as set forth in claim 15 , wherein the measure of similarity (d k ) is calculated by determining a Euclidean distance or an absolute value.
18 . A method as set forth in claim 15 , wherein the images of the second image sequence are calculated by linear spline approximation or a higher-level or polynomial approximation.
19 . A method as set forth in claim 15 , wherein in determining the predominant direction of motion, a vertical mid-region of a current image (x 0 ) of the first image sequence is compared to different vertical regions of a previous image (x 1 ) of this sequence and it is determined whether the vertical region of the previous image with the greatest similarity to the mid-region of the current image is situated left or right of center.
20 . A method as set forth in claim 15 , wherein in determining the predominant direction of motion, a second measure of similarity (d I ) between the image regions is calculated by determination of a Euclidean distance or an absolute value.
21 . A method as set forth in claim 15 , wherein a blurring region (ε) with which small movements can be suppressed is established around the mid-region of the current image (x 0 ).
22 . A system for generating 3-D images from a first sequence of 2-D images, comprising
a first input for receiving the first image sequence; a motion analyzer connected to the first input for comparing a pair of sequential images in the first image sequence to determine a measure of similarity (d k ) between the pair of sequential images and for setting an approximation variable (a) that determines a stereo base based on the measure of similarity; an image generator connected to the motion analyzer for generating a synthesized image of a second sequence of images by interpolating between the pair of images of the first image sequence using the approximation variable; first and second outputs for transmitting the first and second image sequences to left and right viewing channels; and a phase selector connected to the image generator for assigning images received from the phase analyzer to one of the left and the right viewing channels.
23 . A system as set forth in claim 16 , comprising a phase analyzer interconnected between the image generator and the phase selector for determining a predominant direction of motion in sequential images of the first image sequence, wherein the phase selector assigns images to one of the left and right viewing channels based on the predominant direction of motion determined by the phase analyzer.
24 . A system as set forth in claim 16 , comprising an analog-to-digital converter for converting analog image data received at the first input to digital image data.
25 . A system as set forth in claim 18 , comprising one or more of (a) a second input connected to the image generator for receiving manual motion control data; and (b) a third input connected to the phase selector for receiving manual phase control data.Join the waitlist — get patent alerts
Track US2007269136A1 — get alerts on status changes and closely related new filings.
We store only your email — no account needed. See our privacy policy.