Color gradient capture from source image content

Patent metadata

Field	Value
Publication number	US-2022058841-A1
Application number	US-202117519623-A
Country	US
Kind code	A1
Filing date	Nov 5, 2021
Priority date	Nov 18, 2019
Publication date	Feb 24, 2022
Grant date	—

Abstract

A method for generating a color gradient includes receiving an input indicating a smoothness of the color gradient and detecting a gradient path defined from an image. The method also includes identifying a set of colors from the gradient path. The method includes detecting a set of color pivots associated with the set of colors. A number of the color pivots in the set of color pivots is based on the input indicating the smoothness of the color gradient. The method includes generating a set of individual color gradients along the gradient path including a color gradient between a first pair of colors located at a first pair of the color pivots and a different color gradient between a second pair of colors located at a second pair of the color pivots. Additionally, the method includes generating the color gradient of the image from the set of individual color gradients.

First claim

1 . A method comprising:
generating a set of potential gradient paths for a source image, wherein each potential gradient path of the set of potential gradient paths is a diagonal line traversing through the source image; determining, for each potential gradient path of the set of potential gradient paths, a gradient goodness score for the potential gradient path, wherein the gradient goodness score indicates an estimated measure of color variance along the potential gradient path; selecting, from the set of potential gradient paths, a potential gradient path having a highest gradient goodness score; determining the selected potential gradient path as a gradient path that represents a color gradient of the source image; and identifying a set of colors from the gradient path to determine the color gradient of the source image.

2 . The method of claim 1 , wherein determining the gradient goodness score for the potential gradient path includes:
determining a total local variance score of the potential gradient path; determining a line variance score of the potential gradient path, wherein the line variance score; and determining the gradient goodness score of the potential gradient path based on a ratio between the total local variance score and the line variance score.

3 . The method of claim 2 , wherein determining the total local variance score of the potential gradient path includes:
identifying a set of points extending along a length of the potential gradient path; determining, for each point of the set of points, a local variance score, wherein the local variance score is calculated based on color values of a line of pixels that: (i) includes a pixel representing the point; and (ii) is formed perpendicular from the potential gradient path; and determining the total local variance score based on the local variance scores of the set of points.

4 . The method of claim 2 , wherein determining the line variance score of the potential gradient path includes:
identifying a red-color representation value for pixels of the potential gradient path; identifying a green-color representation value for the pixels of the potential gradient path; identifying a blue-color representation value for the pixels of the potential gradient path; and determining the line variance score based on the red-color representation value, the green-color representation value, and the blue-color representation value of the potential gradient path.

5 . The method of claim 1 , further comprising modifying the determined color gradient of the source image by applying a color pivot between a first color and a second color of the set of colors.

6 . The method of claim 5 , wherein modifying the determined color gradient of the source image includes increasing a smoothness of the color gradient, wherein the smoothness of the color gradient is increased by reducing a number of color pivots being applied between colors of the set of colors.

7 . The method of claim 5 , wherein modifying the determined color gradient of the source image includes decreasing a smoothness of the color gradient, wherein the smoothness of the color gradient is decreased by increasing a number of color pivots being applied between colors of the set of colors.

8 . A system comprising:
a processing device; and a memory device communicatively coupled to the processing device and comprising program instructions, wherein when executed, cause the processing device to perform operations comprising:
generating a set of potential gradient paths for a source image, wherein each potential gradient path of the set of potential gradient paths is a diagonal line traversing through the source image;
determining, for each potential gradient path of the set of potential gradient paths, a gradient goodness score for the potential gradient path, wherein the gradient goodness score indicates an estimated measure of color variance along the potential gradient path;
selecting, from the set of potential gradient paths, a potential gradient path having a highest gradient goodness score;
determining the selected potential gradient path as a gradient path that represents a color gradient of the source image; and
identifying a set of colors from the gradient path to determine the color gradient of the source image.

9 . The system of claim 8 , wherein determining the gradient goodness score for the potential gradient path includes:
determining a total local variance score of the potential gradient path; determining a line variance score of the potential gradient path, wherein the line variance score; and determining the gradient goodness score of the potential gradient path based on a ratio between the total local variance score and the line variance score.

10 . The system of claim 9 , wherein determining the total local variance score of the potential gradient path includes:
identifying a set of points extending along a length of the potential gradient path; determining, for each point of the set of points, a local variance score, wherein the local variance score is calculated based on color values of a line of pixels that: (i) includes a pixel representing the point; and (ii) is formed perpendicular from the potential gradient path; and determining the total local variance score based on the local variance scores of the set of points.

11 . The system of claim 9 , wherein determining the line variance score of the potential gradient path includes:
identifying a red-color representation value for pixels of the potential gradient path; identifying a green-color representation value for the pixels of the potential gradient path; identifying a blue-color representation value for the pixels of the potential gradient path; and determining the line variance score based on the red-color representation value, the green-color representation value, and the blue-color representation value of the potential gradient path.

12 . The system of claim 8 , wherein the program instructions further cause the processing device to perform operations comprising:
modifying the determined color gradient of the source image by applying a color pivot between a first color and a second color of the set of colors.

13 . The system of claim 12 , wherein modifying the determined color gradient of the source image includes increasing a smoothness of the color gradient, wherein the smoothness of the color gradient is increased by reducing a number of color pivots being applied between colors of the set of colors.

14 . The system of claim 12 , wherein modifying the determined color gradient of the source image includes decreasing a smoothness of the color gradient, wherein the smoothness of the color gradient is decreased by increasing a number of color pivots being applied between colors of the set of colors.

15 . A non-transitory computer-readable medium storing program instructions that, when executed by one or more processing devices, cause the one or more processing devices to perform operations comprising:
generating a set of potential gradient paths for a source image, wherein each potential gradient path of the set of potential gradient paths is a diagonal line traversing through the source image; determining, for each potential gradient path of the set of potential gradient paths, a gradient goodness score for the potential gradient path, wherein the gradient goodness score indicates an estimated measure of color variance along the potential gradient path; selecting, from the set of potential gradient paths, a potential gradient path having a highest gradient goodness scor

Assignees

• Adobe Inc

Inventors

• Biswas Mainak
• Diverdi Stephen Joseph
• Vallespi Jose Ignacio Echevarria

Classifications

• G06T11/10 Primary

  Texturing; Colouring; Generation of textures or colours (retouching, inpainting or scratch removal G06T5/77) · CPC title

• G06T2200/24

  involving graphical user interfaces [GUIs] · CPC title

• G06T2207/10024

  Color image · CPC title

• G06T7/90

  Determination of colour characteristics · CPC title

• G06T11/001 Primary

  Physics · mapped topic