Method and device for the creation of pseudo-holographic images
A method and a device enable the creation of three-dimensional images with more than two perspectives (e.g. pseudo-holographic images), especially to be reproduced with the aid of an autostereoscopic display or an autostereoscopic screen, from fed images having, in particular, only two perspectives, e.g., a left and a right image channel. Also, a related device creates and reproduces three-dimensional images having more than two perspectives, especially in the form of an autostereoscopic multi-user visualization system.
1. A method for displaying three-dimensional images having a number N>2 perspective on an autostereoscopic display or an autostereoscopic screen, comprising:
receiving fed images with a number M larger or equal to 2 perspectives, where M is less than N;
analyzing each pixel B(i,j) of a fed image of a first perspective to determine a corresponding pixel B(i′,j′) in a second perspective or a right occultation, where i and i′ are row indices and j and j′ are column indices of the fed images;
associating each pixel or subpixel of the display or screen with one of the M fed image perspectives or one of N minus M in-between perspectives;
synthesizing the pixels or subpixels of the N minus M in-between perspectives associated with a pixel or subpixel of the display or screen through a virtual camera movement along a stereo basis between a pixel B(i,j) of the fed image of the first perspective and the corresponding pixel B(i′,j′) of the fed image of the second perspective, without synthesizing at least one of the pixels or subpixels of the N minus M in-between perspectives which is not associated with a pixel or subpixel of the display or screen; and
displaying each associated pixel or subpixel of the display or screen on to the display or screen.
2. The method according to claim 1 , wherein the fed images M=2 are a left image and a right image.
3. The method according to claim 2 , further comprising choosing α k supporting points with 0<α k <1 for k=1, . . . N−2 for the creation of N minus 2 in-between perspectives, where the pixels B(i,j(α k )) are to be combined to an in-between perspective.
4. A device for creating three-dimensional images having a number N>2 perspectives for visualization on an autostereoscopic display or an autostereoscopic screen, comprising:
an input for receiving fed images with a number M larger or equal to 2 perspectives, where M is less than N;
an analysis unit for analyzing each pixel B(i,j) of a fed image of a first perspective to determine a corresponding pixel B(i′,j′) in a second perspective or a right occultation, where i and i′ are row indices and j and j′ are column indices of the fed images;
a data structure containing data associating each pixel or subpixel of the display or screen with one of the M fed image perspectives or one of N minus M in-between perspectives; and
an image synthesis unit for synthesizing the pixels or subpixels of the N minus M in-between perspectives associated with a pixel or subpixel of the display or screen through a virtual camera movement along a stereo basis between a pixel B(i,j) of the fed image of the first perspective and the corresponding pixel B(i′,j′) of the fed image of the second perspective, without synthesizing at least one of the pixels or subpixels of the N minus M in-between perspectives which is not associated with a pixel or subpixel of the display or screen.
5. The device according to claim 4 , wherein:
the analysis unit generates, for each pixel B(i,j) of the fed image of the first perspective an occultation indicator and a whereto pointer;
where the occultation indicator indicates whether there is a corresponding pixel B(i,j′) in the fed image of a second perspective for the pixel B(i,j) in the fed image of the first perspective or, whether there is a right occultation or a left occultation; and
where the whereto pointer holds, if the occulator occultation indicator indicates there is a corresponding pixel where i and i′ are row indices and j and j′ are column indices of the fed images.
6. A computer program product, the computer program product comprising:
a non-transient storage medium storing computer readable instructions;
the computer readable instructions including instructions for:
receiving fed images with a number M larger or equal to 2 perspectives, where M is less than N;
analyzing each pixel B(i,j) of a fed image of a first perspective to determine either a corresponding pixel (i′,j′) in a second perspective or a right occultation, where i and i′ are row indices and j and j′ are column indices of the fed images;
associating each pixel or subpixel of an autostereoscopic display or an autostereoscopic screen with one of the M fed image perspectives or one of N minus M in-between perspectives;
synthesizing the pixels or subpixels of the N minus M in-between perspectives associated with a pixel or subpixel of the display or screen through a virtual camera movement along a stereo basis between a pixel B(i,j) of the fed image of the first perspective and the corresponding pixel B(i′,j′) of the fed image of the second perspective, without synthesizing at least one of the pixels or subpixels of the N minus M in-between perspectives which is not associated with a pixel or subpixel of the display or screen; and
transmitting each associated pixel or subpixel of the display or screen to the display or screen.
7. The method according to claim 2 , further comprising during the synthesis step synthesizing a whereto pointer which identifies the corresponding pixel B(i′,j′) in the second perspective.
8. The method according to claim 7 , further comprising a correspondence analysis by a transformation of a transmitted depth map.
9. The method according to claim 1 , further comprising a step of receiving the number N from a user.
10. The device according to claim 4 , wherein the analysis unit generates for each pixel B(i,j) of the fed image of the first perspective a whereto pointer where the whereto pointer (i,j)=j′ when there is a corresponding pixel B(i,j′) of the fed image of the second perspective.
11. The method according to claim 1 , wherein the step of associating each pixel or subpixel of the display or screen with one of the M fed image perspectives or one of N minus M in-between perspectives comprises a matrix R(i,j) which identifies the perspective for each pixel or subpixel of the display or screen.
12. The method according to claim 1 , further comprising receiving fed correspondences between pixels B(i,j) of the fed image of the first perspective and pixels B(i′,j′) of the fed image of the second perspective.
13. The method according to claim 12 , wherein the step of analysing analyzing comprises determining a corresponding pixel B(i,j) in the second perspective from the fed correspondences.
14. The method according to claim 1 , wherein determining a corresponding pixel B(i,j) in the second perspective comprises a search process.