Sample based color extraction for augmented reality
A system and method for sampling-based color extraction for augmented reality are described. A viewing device includes an optical sensor to capture an image of a real-world object. A color extraction software divides the captured image into multiple regions or recognizes pre-defined regions and identifies a color value for each region. A color-based augmented reality effect module retrieves a virtual content based on the color values for the regions, and delivers the virtual content in the viewing device.
1. A method comprising:
capturing, by an optical sensor of a viewing device, an image of a real-world object, yielding image data;
identifying, based on the image data, the real-world object depicted in the image;
processing the image data using a processor of the viewing device to divide the captured image of the real-world object into at least a first region and a second region;
identifying a color value for the first region and a color value for the second region;
identifying a color value variation for the first region;
determining that the color value variation for the first region is greater than an upper threshold for a desired color value variation range;
in response to determining that the color value variation for the first region is greater than the upper threshold, increasing a size of the first region;
retrieving virtual content data based on an identification of the real-world object, the color value for the first region, and the color value for the second; and
generating a visualization of virtual content from the virtual content data in a display of the viewing device, the virtual content including colors indicated by the color value for the first region and the color value for the second region.
2. The method of claim 1 , further comprising:
identifying regions in the captured image having sample pixel color values within a predefined threshold; and
dividing the captured image into a plurality of regions based on the identified regions.
3. The method of claim 2 , further comprising:
identifying a color value variation for the second region;
determining that the color value variation for the second region is less than an lower threshold for the desired color value variation range; and
in response to determining that the color value variation for the second region is less than the lower threshold, decreasing a size of the second region.
4. The method of claim 1 , further comprising:
computing a median color value for the first region.
5. The method of claim 4 , wherein computing the median color value for the first region comprises:
selecting sample pixels from within the first region; and
computing the median color value based on the sample pixels from within the first region.
6. The method of claim 1 , wherein
the virtual content is a three-dimensional model that when virtualized in the display of the viewing device is visually perceived as an overlay on top of the captured image.
7. The method of claim 6 , further comprising:
retrieving a characteristic of the virtual content; and
rendering a visualization of the characteristic of the virtual content in the display of the viewing device.
8. The method of claim 7 , wherein the characteristic includes an animation or a color of the virtual content, a first animation of the virtual object associated with a first color value, a second animation of the virtual object associated with a second color value, a first color of the virtual object associated with the first color value a second color of the virtual object associated with the second color value.
9. The method of claim 1 ,
wherein the visualization of the virtual content is visually perceived as an overlay to the real-world object through the display of the viewing device, the display being a transparent display.
10. A non-transitory machine-readable medium comprising instructions that, when executed by one or more computer processors of a viewing device, cause the viewing device to perform operations comprising:
capturing, by an optical sensor of the viewing device, an image of a real-world object, yielding image data;
identifying, based on the image data, the real-world object depicted in the image;
processing the image data to divide the captured image of the real-world object into at least a first region and a second region;
identifying a color value for the first region and a color value for the second region;
identifying a color value variation for the first region;
determining that the color value variation for the first region is greater than an upper threshold for a desired color value variation range;
in response to determining that the color value variation for the first region is greater than the upper threshold, increasing a size of the first region;
retrieving virtual content data based on an identification of the real-world object the color value for the first region, and the color value for the second region; and
generating a visualization of virtual content from the virtual content data in a display of the viewing device, the virtual content including colors indicated by the color value for the first region and the color value for the second region.
11. A viewing device comprising:
one or more computer processors; and
one or more computer-readable mediums storing instructions that, when executed by the one or more computer processors, cause the viewing device to perform operations comprising:
capturing, with an optical sensor of the viewing device, an image of a real-world object, yielding image data;
identifying, based on the image data, the real-world object depicted in the image;
processing the image data to divide the captured image of the real-world object into at least a first region and a second region;
identifying a color value for the first region and a color value for the second region;
identifying a color value variation for the first region;
determining that the color value variation for the first region is greater than an upper threshold for a desired color value variation range;
in response to determining that the color value variation for the first region is greater than the upper threshold, increasing a size of the first region;
retrieving virtual content data based on an identification of the real-world object, the color value for the first region, and the color value for the second region; and
generating a visualization of virtual content from the virtual content data in a display of the viewing device, the virtual content including colors indicated by the color value for the first region and the color value for the second region.
12. The viewing device of claim 11 , the operations further comprising:
identifying regions in the captured image having sample pixel color values within a predefined threshold; and
dividing the captured image into a plurality of regions based on the identified regions.
13. The viewing device of claim 12 , the operations further comprising:
identifying a color value variation for the second region;
determining that the color value variation for the second region is less than an lower threshold for the desired color value variation range; and
in response to determining that the color value variation for the second region is less than the lower threshold, decreasing a size of the second region.
14. The viewing device of claim 11 , the operations further comprising:
computing a median color value for the first region.
15. The viewing device of claim 14 , wherein computing the median color value for the first region comprises:
selecting sample pixels from within the first region; and
computing the median color value based on the sample pixels from within the first region.
16. The viewing device of claim 11 , wherein the virtual content is a three-dimensional model that when virtualized in the display of the viewing device is visually perceived as an overlay on top of the captured image.
17. The viewing device of claim 16 , the operations further comprising:
retrieving a characteristic of the virtual content; and
rendering a visualization of the characteristic of the virtual content in the display of the viewing device.
18. The viewing device of claim 17 , wherein the characteristic includes an animation or a color of the virtual content, a first animation of the virtual object associated with a first color value, a second animation of the virtual object associated with a second color value, a first color of the virtual object associated with the first color value, a second color of the virtual object associated with the second color value.
19. The viewing device of claim 11 , wherein the visualization of the virtual content is visually perceived as an overlay to the real-world object through the display of the viewing device, the display being a transparent display.