Image analysis – Image transformation or preprocessing – Mapping 2-d image onto a 3-d surface
Reexamination Certificate
1998-12-29
2001-11-06
Couso, Yon J. (Department: 2723)
Image analysis
Image transformation or preprocessing
Mapping 2-d image onto a 3-d surface
C348S042000
Reexamination Certificate
active
06314211
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to an apparatus and a method for generating a three-dimensional image, and more particularly, to an apparatus and a method for converting a two-dimensional image sequence into a three-dimensional image using the motion of the two-dimensional image sequence and post-processing method during the generation of a three-dimensional image.
2. Description of the Related Art
FIG. 1
is a flowchart for describing a conventional method for generating a three-dimensional image using a modified time difference.
Sanyo Electronics Co. Ltd. has developed a method using a modified time difference (MTD) in the conventional method for generating the three-dimensional image. Referring to
FIG. 1
, a moved region is extracted from the two dimensional image sequence (step
100
). The moving speed and direction of the extracted area are extracted (step
110
). A delay direction and a delay time are determined from a previous image frame based on the moving speed and direction of a current image (step
120
). Then, a determination is made as to which eye a delayed image and the current image are respectively displayed according to the determined delay direction and delay time (step
130
).
FIG. 2
describes processes of determining left/right images in the method described in FIG.
1
. To be more specific,
FIG. 2
describes processes of determining the left/right images when a camera for capturing the images of the two-dimensional image sequence is fixed and a subject ▪ in an image moves. According to the conventional method using the MTD, a binocular disparity is provided to a person by showing the image time delayed in the left eye shown in
FIG. 2. A
binocular disparity opposite to the above-mentioned one is provided by showing an original image to the left eye and the time delayed image to the right eye when the subject ▪ moves to the left.
The conventional method for generating the three-dimensional image using the MTD has the following problems.
First, since an image is selected as the delayed image among previous frames based on only motion information of the original image, other depths of the respective regions of the original image are ignored. As a result, a sense of depth cannot be felt and a cubic effect can be felt in a moving subject. However, it is not possible to feel the cubic effect in a subject which barely moves such as the background of a screen.
Second, when the motion of a subject has a vertical component between the original image and the delayed image, if the three-dimensional image is generated by the conventional method using the MTD, the viewer's eyes become tired due to the visual structure of humans in which a sense of depth is obtained by a horizontal disparity.
Third, when the motion vector of a subject in the original image is very large, even if a frame is selected as the delayed image from a plurality of previous frames, the disparity of the stereo-image generated by the original image and the delayed image is beyond the limits of humans to perceive disparity.
In general, in the method for generating the three-dimensional image from the two dimensional image sequence, the two-dimensional image sequence photographed by a single eyed video camera and another image corresponding thereto are used. Here, how another image is made is important. If the quality of the made image is deteriorated, for example, the real motion of the subject in a single eyed image is not described well and the boundary of the subject is broken, a natural three-dimensional image is not generated.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide an apparatus for converting a two dimensional image sequence into a three dimensional image using a conversion of a motion disparity into a horizontal disparity by which it is possible to generate a new image by measuring the motion of the two-dimensional image sequence and converting the motion into the horizontal disparity according to the characteristics of the motion, thus generating a three-dimensional image in which a sense of depth is felt in each region of the image, a natural cubic effect is felt even if the motion of a subject in an original image has a vertical component, and the motion vector of the subject in the original image can be perceived by the human disparity no matter how large it is.
It is another object of the present invention to provide a three-dimensional image converting method performed by the three-dimensional image converting apparatus.
It is still another object of the present invention to provide a post processing method during generation of the three-dimensional image using the two-dimensional image sequence by which it is possible to improve the quality of a disparity image and to generate a natural stereo-image by discriminating the presence of the same subjects in adjacent blocks using brightness characteristics for the respective blocks in the original image have and post-processing the disparity image according to the discrimination result.
Accordingly, to achieve the first object, there is provided an apparatus for converting a two-dimensional image sequence into a three-dimensional image, comprising a block motion measuring portion for measuring a motion vector for each block of a current image divided into blocks having a predetermined size using a previous image frame, a horizontal disparity generating portion for obtaining a horizontal disparity from the motion vector of each block according to the motion characteristic of the current image, an image generating portion for moving each block in a horizontal direction according to each horizontal disparity and generating a composite image corresponding to the current image, and an outputting portion for displaying a three-dimensional image comprised of the current image and the composite image.
To achieve the second object, there is provided a method for converting a two-dimensional image sequence into a three-dimensional image, comprising the steps of (a) dividing an input current image into blocks having a predetermined size, (b) measuring a motion vector for each block of the current image using a previous image frame, (c) obtaining a horizontal disparity from the motion vector of each block according to the motion characteristic of the current image, (d) generating a composite image corresponding to the current image by moving each block in a horizontal direction according to the each horizontal disparity, and (e) displaying a three-dimensional image formed of the current image and the composite image.
To achieve the third object, there is provided a method for post-processing a displacement image made by extracting the motion of an original image in a system for generating a stereo image using a single eyed image sequence, comprising the steps of (a) obtaining the disparity of each block of the original image divided into blocks having a predetermined size using the disparity image and a maximum disparity and generating a disparity map comprised of disparities of all the blocks, (b) comparing brightness characteristics between each block and vertically adjacent blocks and determining the group of block having similar brightness characteristics to be the filtering mask, (c) replacing the disparity of each block by a new disparity obtained from blocks in the concerned filtering mask and modifying the disparity map, and (d) reconstructing the disparity image using the original image and the modified disparity map.
REFERENCES:
patent: 5619256 (1997-04-01), Haskell et al.
patent: 5682437 (1997-10-01), Okino et al.
patent: 5808664 (1998-09-01), Yamashita et al.
patent: 6043838 (2000-03-01), Chen
patent: HEI 8-289328 (1996-11-01), None
patent: 9-81746 (1997-03-01), None
patent: 97-58060 (1997-07-01), None
Kim Do-kyoon
Kim Man-bae
Song Mun-sup
Burns Doane Swecker & Mathis L.L.P.
Couso Yon J.
Samsung Electronics Co,. Ltd.
LandOfFree
Apparatus and method for converting two-dimensional image... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and method for converting two-dimensional image..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and method for converting two-dimensional image... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2613526