Systems and methods for color palette suggestions

• US 10,235,424 B2
• Filed: 05/01/2015
• Issued: 03/19/2019
• Est. Priority Date: 05/06/2014
• Status: Active Grant

• Current Assignee: Shutterstock Incorporated
• Original Assignee: Shutterstock Incorporated
• Inventors: Becker, Chris, Kroner, David, Blaurock, Steve
• Primary Examiner(s): Gebresenbet, Dinku W
• Application Number:
  US14/701,547
• Publication Number:
  US 20150324366A1
• Time in Patent Office: 1,418 Days
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06F 16/5866 using information manually generated, e.g. tags, keywords, comments, manually generated location and time information

Abstract

A method and system for conducting image search is provided that includes searching a first database to locate a set of pertinent images, iteratively performing the following operations for each image in the set of pertinent images (a) extracting the histogram of red green and blue colors (RGB colors) from a given image (b) distilling the extracted RGB colors down to create a reduced color palette for the given image (c) segmenting the extracted RGB colors into a set of segments representing distinct parts of the color spectrum; (d) selecting a subset of segments to assemble a color palette for the given image; (e) updating the assembled color palette and the customer behavior score for the given image in the first database; and generating a ranked display of suggested color palettes for the search query by aggregating the customer behavior scores for the search query across all images.

Claims

1. A data processing system for recommending a color palette comprising:

a data processing unit to execute steps to;

receive from a contributing user an image and at least one keyword;

extract at least one color from the image to provide at least one extracted color;

divide the at least one extracted color into reduced colors associated with visually distinct parts of a color spectrum, each of the reduced colors associated with a bucket;

calculate a bucket representation score for the at least one extracted color by summing bucket color scores of each bucket of the image;

generate a palette by assigning to the palette at least one reduced color from among the reduced colors, wherein the at least one reduced color is the reduced color having the highest bucket representation score for the at least one extracted color;

determine a customer behavior score for the image based on aggregated activity for the image;

obtain a color weight score for each reduced color in the image based on other visually similar colors to the respective reduced color in the image;

obtain a color volume score for each reduced color based on a percentage of the image covered by the reduced color;

index the customer behavior score, color weight score, and color volume score for a search engine;

receive from a searching user a search query comprising a search term comprising the at least one keyword; and

provide for display, a grid of search results comprising the image displayed according to the customer behavior score, the at least one extracted color from the image when the searching user interacts with the image, and at least one palette comprising the reduced color among colors ranked according to a customer behavior score associated with the at least one keyword,

wherein the color weight score for each reduced color is calculated based on a percentage of the respective reduced color in the image, a percentage of other colors within a predefined distance of the respective reduced color in a red, green, and blue color space, and a measure of distance between the respective reduced color and the other colors within a predefined distance of the respective reduced color in a red, green, and blue color space.

2. The system of claim 1, wherein the customer behavior score is calculated based on user activity, the user activity including searches which lead to a purchase of the image, clicking activity on the image, or other signals which would indicate user interest in the image, and wherein the customer behavior score is weighted based on a specific period of a year.

3. The system of claim 1, wherein the colors of the at least one palette further comprise at least one color present among all images in the search results, and wherein the colors of the at least one palette are sorted by customer behavior score for the at least one keyword, and wherein the processor is further configured to permit the searching user to create another palette using the at least one color present among all images in the search results.

4. The system of claim 1, wherein the search results are ranked based on proximity of volume of each color of each image of said search results to volume of each extracted color in a selected image.

5. The system of claim 1, wherein the data processing unit is further configured to execute steps to:

display a second palette containing at least one extracted color from an image when said user hovers a cursor over the image;

display a third palette with a list of at least one color selected by a user; and

display one other color in the third palette.

6. The system of claim 1, wherein the data processing unit is further configured to execute steps to receive from the searching user a request to apply a previously-created palette to a future search query.

7. The system of claim 1, wherein the data processing unit is further configured to execute steps to:

define the buckets manually by segmenting the color spectrum into sections that are visually distinct from one another;

assign a bucket color score to each of the buckets, wherein the bucket color score is calculated by extracting the lightness, chroma, and hue values of the color associated with the respective bucket, and multiplying the lightness value by the chroma value and by the color volume score.

8. The system of claim 1, wherein the data processing unit is further configured to execute steps to generate a data set which contains a list of at least one color palette ranked by aggregated customer behavior scores.

9. The system of claim 1, wherein the data processing unit is further configured to execute steps to receive input of a search term and a color from the searching user and update the search results to display at least one image which matches the search term and said color.

10. A method for executing a data processing application for recommending a color palette, the method comprising the steps of:

receiving from a contributing user an image and at least one keyword;

extracting at least one color from said image to provide at least one extracted color;

dividing said at least one extracted color of said image into reduced colors associated with visually distinct parts of a color spectrum, each of the reduced colors associated with a bucket;

calculating a bucket representation score for the at least one extracted color by summing bucket color scores of each bucket of said image;

generating a palette by assigning to said palette at least one reduced color from among the reduced colors, wherein the at least one reduced color is the reduced color having the highest bucket representation score for the at least one extracted color;

determining a customer behavior score for the image based on aggregated activity for the image;

obtaining a color weight score for each reduced color in said image based on other visually similar colors to the respective reduced color in said image;

obtaining a color volume score for each reduced color based on a percentage of said image covered by said reduced color;

indexing the customer behavior score, color weight score, and color volume score for a search engine;

receiving from a searching user a search query comprising a search term comprising the at least one keyword; and

provide for display, a grid of search results comprising the image displayed according to the customer behavior score, the at least one extracted color from the when the searching user interacts with the image, and at least one palette ranked according to a customer behavior score associated with the least one keyword,

wherein the color weight score for each reduced color is calculated based on a percentage of the respective reduced color in the image, a percentage of other colors within a predefined distance of the respective reduced color in a red, green, and blue color space, and a measure of distance between the respective reduced color and the other colors within a predefined distance of the respective reduced color in a red, green, and blue color space.

11. The method of claim 10, wherein the customer behavior score is calculated based on user activity, the user activity including searches which lead to a purchase of the image, clicking activity on the image, or other signals which would indicate interest in the image, and wherein the customer behavior score is weighted based on a specific period of a year.

12. The method of claim 10, wherein the colors of the at least one palette further comprise at least one color present among all images in the search results, wherein the colors of the at least one palette are sorted by customer behavior score for the at least one keyword, and wherein the method further comprises permitting the searching user to create another palette using the at least one color present among all images in the search results.

13. The method of claim 10, wherein said application ranks the search results are ranked based on proximity of volume of each color of each image of the search results to volume of each extracted color in a selected image.

14. The method of claim 10, further comprising:

displaying a second palette containing at least one extracted color from an image when the user hovers a cursor over the image;

displaying a third palette with a list of at least one color selected by the user; and

displaying one other color in the third palette.

15. The method of claim 10, further comprising receiving from the searching user a request to apply a previously-created palette to a future search query.

16. The method of claim 10, further comprising:

defining the bucket manually by segmenting the color spectrum into sections that are visually distinct from one another;

assigning a bucket color score to each of the buckets, wherein the bucket color score is calculated by extracting lightness, chroma, and hue values of the color associated with the respective bucket, and multiplying the lightness value by the chroma value and by the color volume score.

17. The method of claim 10, further comprising generating a data set which contains a list of at least one color palette ranked by aggregated customer behavior score.

18. The method of claim 10, further comprising receiving input of a search term and a color from the searching user and updating the list to display at least one image which matches the search term and the color.