Synthetic panoramagram

• US 6,366,281 B1
• Filed: 09/13/1999
• Issued: 04/02/2002
• Est. Priority Date: 12/06/1996
• Status: Expired due to Term

• Current Assignee: RealD Incorporated
• Original Assignee: Stereographics Corporation (RealD Incorporated)
• Inventors: Milici, Dave, Andersen, Steve, Lipton, Lenny
• Primary Examiner(s): Nguyen, Phu K.
• Application Number:
  US09/319,428
• Time in Patent Office: 932 Days
• Field of Search: 345/419, 345/427, 345/426, 345/428, 345/606, 345/607
• US Class Current: 345/419
• CPC Class Codes: G03C 9/02 Parallax-stereogram
• CPC Class Codes: G06T 13/80 2D [Two Dimensional] animation, e.g. using sprites
• CPC Class Codes: G06T 2207/10012 Stereo images
• CPC Class Codes: G06T 2210/44 Morphing
• CPC Class Codes: G06V 10/7515 Shifting the patterns to accommodate for positional errors
• CPC Class Codes: H04N 13/161 Encoding, multiplexing or demultiplexing different image signal components for multi-view video sequence encoding H04N19/597
• CPC Class Codes: H04N 13/275 from 3D object models, e.g. computer-generated stereoscopic image signals
• CPC Class Codes: H04N 13/282 for generating image signals corresponding to three or more geometrical viewpoints, e.g. multi-view systems
• CPC Class Codes: H04N 13/305 using lenticular lenses, e.g. arrangements of cylindrical lenses

Abstract

A method for synthesizing multiple intermediate view images from a pair of planostereoscopic source images. The source images have different perspective views which are spaced apart in a horizontal plane. Control points are defined at corresponding locations in the source images, and each control point includes position and color information. Intermediate images are created as transformations of the source images by “morphing” from one source image to the other using position and color information from each of the corresponding locations. In a panoramagram embodiment, the intermediate images and the source images are then interdigitated to create a single output image with a continuum of views ranging from one source image to the other source image.

Claims

1. A method for making multiple view images for an autostereoscopic display, comprising generating a plurality of intermediate images from a pair of source images, wherein each source image is defined by multiple points each having position and color information associated therewith and a view perspective which is spaced apart from the other image, wherein each of the plurality of intermediate images has a different view perspective between the source images, and wherein each intermediate image is a transformation of the source images based on the position and color information at multiple points of the source images.

2. The method of claim 1, wherein the generating step comprises:

a. creating a first plurality of control points for a first working area of one source image, each control point having unique position and color information associated therewith,

b. creating a second plurality of control points for a corresponding second working area of the other source image, each control point having unique position and color information associated therewith,

c. repeating steps a and b for a plurality of working areas in each source image, and

d. creating a plurality of transformations of the source images each as a function of the position and color information for each control point.

3. A method for interdigitating multiple view images, wherein each view image is defined by a matrix having rows and columns of pixels, comprising:

scanning each view image one row at a time;

storing a pixel from each column in a respective vertical buffer during a scan step, and

combining the vertical buffers into a single output image.

4. A method for making a panoramagram comprising generating a plurality of intermediate images from a pair of planostereoscopic source images each having position and color information and a view perspective which is spaced apart from the other image, each of the intermediate images having a different view perspective which is between the source images and is a transformation of the source images based on the position and color information of the source images, then interdigitating the intermediate images and the source images to form a single output image having substantially a continuum of views ranging from one source image to the other source image.

5. A method for making a panoramagram, comprising:

preparing a pair of planostereoscopic source images, namely a left image and right image each having position and color information and a view perspective which is spaced apart from the other image,

generating a plurality of intermediate images each having a different view perspective which is intermediate to that of the source images and is based on the position and color information from the source images, and

interdigitating the intermediate images and the source images to create a single output image having substantially a continuum of views ranging from the left image to the right image.

6. The method of claim 5, wherein the generating step comprises:

a. creating a first plurality of control points for a first working area of one source image, each control point having unique position and color information,

b. creating a second plurality of control points for a corresponding second working area of the other source image, each control point having unique position and color information,

c. repeating steps a and b for a plurality of working areas in each source image, and

d. creating a plurality of intermediate images which are transformations of the source images, each of said transformations being a function of the position and color information for each control point.

7. The method of claim 5, wherein the generating step comprises:

a. creating a first control point for a first working area of one source image, said first control point having unique position and color information,

b. creating a last control point for a second working area of the other source image, wherein the second working area of the other source image substantially corresponds in position to the first working area of the one source image, said last control point having unique position and color information,

c. repeating steps a and b for a plurality of working areas in each source image, and

d. creating a plurality of intermediate images which are transformations of the source images, each of said transformations being logically related to movement from the first control point to the last control point for each working area.

8. The method of claim 7, wherein each transformation has position and color information which is logically related to the position and color information for each control point in direct proportion to the proximity of each control point to the transformation.

9. The method of claim 5, wherein the interdigitating step comprises:

extracting a vertical slice from each image,

aligning each vertical slice, and

sequentially ordering the vertical slices.

10. The method of claim 9, wherein the extracting step comprises extracting each vertical slice from a corresponding vertical position in each image.

11. The method of claim 9, wherein the extracting step comprises sequentially extracting vertical slices from a selected block of images.

12. A method for making a panoramagram, comprising:

a. converting each of a pair of planostereoscopic source images, namely a left image and right image, to an RGB format, wherein each source image has a view perspective which is spaced apart from the other source image,

b. aligning the source images in a positional sense,

c. shifting the source images relative to each other to achieve a desired parallax between the source images,

d. resizing the source images to have one dimension which is divisible by a desired number of views and another dimension which maintains a desired aspect ratio between the two dimensions,

e. identifying one of the source images as a start frame and the other of the source images as the end frame,

f. selecting the desired number of views,

g. selecting a working area in one of the source images and placing the working area substantially in a center of a working box,

h. creating a plurality of control points disposed around the center of the working box, each of said control points having position and color information,

i. performing steps g and h for the other of the source images,

j. interactively performing steps g through i as required for a plurality of working areas,

k. creating a plurality of synthetic images equal to the desired number of views by morphing the source images using the position and color information from each control point, and

l. interdigitating the synthetic images and the source images to create a single output image having substantially a continuum of views ranging from the left image to the right image.

13. The method of claim 12, wherein the one dimension is integrally divisible by the desired number of views.

14. A autostereoscopic display surface comprising substantially a continuum of perspective views ranging from a first image at one end of the display surface to a second image at another end of the display surface, wherein a plurality of intermediate images are synthetically generated from the first and second images by creating transformations of position and color information from a plurality of control points in corresponding locations on the first and second images, said transformations being directly proportionate to a proximity from each control point.

15. A stereoscopic panoramagram having a display surface and a lenticular screen in intimate juxtaposition therewith, wherein the display surface includes a plurality of columns, each of said columns having a lenticule corresponding thereto, characterized by each of said columns having substantially a continuum of perspective views ranging from a first image at one end of the column to a second image at another end of the column, and wherein a plurality of intermediate images are generated from the first and second images by creating transformations of position and color information from a plurality of control points in corresponding locations on the first and second images, said transformations being directly proportionate to a proximity from each control point.