Pulling keys from color segmented images
View Patent ↗Described are computer-based methods and apparatuses, including computer program products, for pulling keys from color segmented images. Data indicative of a two dimensional image is stored in a data storage device, the two dimensional image comprising a plurality of pixels. A plurality of color segmented frames are generated based on the two dimensional image, wherein each color segmented frame comprises one or more objects. For each of the color segmented frames, a key is generated based on the one or more objects. A depth map is calculated for the two dimensional image based on the keys, wherein the depth map comprises data indicative of three dimensional information for each pixel of the two dimensional image.
1. A computerized method for pulling keys from a plurality of color segmented images, the method comprising:
storing data indicative of a two dimensional image in a data storage device, the two dimensional image comprising a plurality of pixels;
generating, by a color segmentation unit of a computer, a plurality of color segmented frames based on the two dimensional image, wherein each color segmented frame comprises one or more objects;
generating, by the color segmentation unit, for each of the color segmented frames, a key based on the one or more objects; and
calculating, by a depth map unit of the computer, a depth map for the two dimensional image based on the keys, wherein the depth map comprises data indicative of three dimensional information for each pixel of the two dimensional image.
2. The method of claim 1 , further comprising allowing adjustment of Bezier points of a Bezier map of an object.
3. The method of claim 2 , further comprising, for each color segmented frame, defining an edge of each object of the one or more objects, comprising:
automatically calculating a Bezier map of the object, the Bezier map comprising a plurality of Bezier points about the object and a plurality of Bezier curves connecting the plurality of Bezier points;
receiving data indicative of adjusting the Bezier map; and
generating a more-detailed map of the object based on the data, wherein the more-detailed map comprises a plurality of additional points in combination with the plurality of Bezier points.
4. The method of claim 1 , wherein generating the color segmented frames comprises defining a plurality of base colors, wherein each base color is used to generate a color segmented frame.
5. The method of claim 4 , wherein each base color from the plurality of base colors comprises a predefined range of color values, wherein a pixel comprising a color value within the predefined range of color values is associated with a color segmented frame generated based on the base color.
6. The method of claim 5 , comprising:
receiving data indicative of a new range of color values for one or more colors of the plurality of base colors; and
adjusting the predefined range of color values for each of the one or more colors based on the data.
7. The method of claim 4 , wherein the plurality of base colors comprises, for each color of the plurality of base colors, a color pair that includes a light color and a dark color.
8. The method of claim 4 , wherein the plurality of base colors comprises brown and beige.
9. The method of claim 1 , wherein calculating the depth map comprises:
determining a three dimensional representation of an object in the two dimensional image results in a portion of the object coming in view that was not in view in the two dimensional image; and
stretching a background behind the object, a side of the object, or both, to fill in the portion of the object coming into view.
10. The method of claim 1 , wherein calculating the depth map comprises:
determining a three dimensional representation of an object in the two dimensional image results in a portion of the object going out of view that was in view in the two dimensional image; and
shrinking a side of the object to hide the portion of the object going out of view.
11. The method of claim 1 , wherein calculating the depth map comprises calculating the three dimensional information for each pixel based on a key factor of the plurality of color segmented frames.
12. The method of claim 11 , wherein the key factor includes hue, saturation, and lightness of the pixel, and calculating comprises:
determining whether the level of saturation, the level of lightness, or both are a high level or a low level; and
if the level of saturation or lightness is a high level, assigning the pixel a near depth value; or
if the level of saturation or lightness is a low level, assigning the pixel a far depth value.
13. The method of claim 11 , wherein the key factor includes an amount of value change among the pixel and a group of neighboring pixels, and calculating comprises:
determining, based on the value change, whether the pixel is part of a flat plane; and
if the pixel is part of the flat plane, assigning the pixel a far depth value.
14. The method of claim 11 , wherein the key factor includes a position of each of the one or more objects in each color segmented frame, and calculating comprises, for each object:
if the position of the object is a lower position of the color segmented frame, assigning each pixel within the object a near depth value;
if the position of the object is an upper position of the color segmented frame, assigning each pixel within the object a far depth value;
if the position of the object is within a corresponding object of another color segmented frame, assigning each pixel within the object a near depth value; or
if the position of the object is at an edge portion of the color segmented frame, assigning each pixel within the object a far depth value.
15. The method of claim 11 , wherein the key factor includes a ratio of a size of each of the one or more objects in each color segmented frame to a size of the color segmented frame, and calculating comprises, for each object, assigning a depth value to each pixel within the object based on the ratio.
16. The method of claim 11 , wherein the key factor includes information indicative of a position of an object in the two dimensional image and a position of a corresponding object in a serial two dimensional image, and calculating comprises:
determining the position of the object in the two dimensional image is different than the position of the corresponding object in the serial two dimensional image; and
assigning each pixel within the object a near depth value.
17. The method of claim 1 , further comprising:
storing data indicative of a previously generated depth map in the data storage device, the depth map comprising data indicative of three dimensional information for each pixel of a corresponding two dimensional image; and
generating a new depth map based on the previously generate depth map, wherein the new depth map comprises data indicative of a larger range of three dimensional information for each pixel of a corresponding two dimensional image.
18. The method of claim 1 , wherein calculating the depth map comprises applying one or more experience rules to adjust the depth map, each rule being configured to adjust the depth map based on a human perception of one or more objects in the two dimensional image.
19. A system for pulling keys from a plurality of color segmented images, the system comprising,
a data storage device configured to store data indicative of a two dimensional image, the two dimensional image comprising a plurality of pixels;
a color segmentation unit in communication with the data storage device configured to:
generate a plurality of color segmented frames based on the two dimensional image, wherein each color segmented frame comprises one or more objects; and
generate, for each of the color segmented frames, a key based on the one or more objects; and
a depth map unit in communication with the color segmentation unit and the data storage device configured to calculate a depth map for the two dimensional image based on the keys, wherein the depth map comprises data indicative of three dimensional information for each pixel of the two dimensional image.
20. The system of claim 19 , further comprising an input unit configured to:
receive the data indicative of the two dimensional image and to store the data in the data storage device; and
receive data indicative of a previously generate depth map and to store the data indicative of the previously generate depth map in the data storage device.
21. The system of claim 19 , further comprising an edge generation unit configured to generate edge information for each object in the two dimensional image.
22. The system of claim 19 , further comprising a three dimensional experience unit configured to apply one or more experience rules to adjust the depth map, each rule being configured to adjust the depth map based on a human perception of one or more objects in the two dimensional image.
23. A computer program product, tangibly embodied in a computer readable storage medium, the computer program product including instructions being operable to cause a data processing apparatus to:
store data indicative of a two dimensional image in a data storage device, the two dimensional image comprising a plurality of pixels;
generate, by a color segmentation unit of a computer, a plurality of color segmented frames based on the two dimensional image, wherein each color segmented frame comprises one or more objects;
generate, by the color segmentation unit, for each of the color segmented frames, a key based on the one or more objects; and
calculate, by a depth map unit of the computer, a depth map for the two dimensional image based on the keys, wherein the depth map comprises data indicative of three dimensional information for each pixel of the two dimensional image.