AUTOMATIC IMAGE-BASED RECOMMENDATIONS USING A COLOR PALETTE

• US 20150379959A1
• Filed: 06/26/2014
• Published: 12/31/2015
• Est. Priority Date: 06/26/2014
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Dorner, Charles Shearer, Blackburn, Jenny Ann, Galore, Janet Ellen, Haitani, Robert Yuji, Ong, Timothy Andrew, Ramos, Gonzalo Alberto, Sayre, Paul Barnhart III, Hazlewood, William R., Pham, Dominick
• Granted Patent: US 9,679,532 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06Q 30/0627 using item specifications
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G09G 5/06 using colour palettes, e.g. look-up tables

Abstract

Systems and methods are described that recommend images, items, and/or metadata based at least in part on a reference color palette or reference color name. A color name can be converted into a representation of the color name in a color space. The reference color can be used to identify images that contain the reference color. The identified images and associated metadata can be analyzed, sorted and provided as an ordered list of items. Systems and methods are also described that identify items that contain colors affiliated with the reference color. Systems and methods are also described that validate color identifier information in metadata associated with an image. Systems and methods are also described that identify non-color specific keywords associated with the reference color.

Claims

1. A computer-implemented method for generating a grouping of recommended items, the computer-implemented method comprising:

under control of a hardware computing device configured with specific computer executable instructions,

obtaining a reference color palette;

determining one or more search colors based at least in part on the reference color palette;

obtaining one or more color images;

identifying one or more matching images from the one or more color images, each matching image containing at least one color that corresponds to at least one of the one or more search colors;

obtaining metadata associated with each matching image of the one or more matching images; and

generating a grouping of recommended items based at least in part on an analysis of the metadata associated with each matching image of the one or more matching images.

2. The computer-implemented method of claim 1 further comprising determining a search color range for each search color of the one or more search colors, the search color range for each search color based at least in part on a color distance from the search color in a color space.

3. The computer-implemented method of claim 2, wherein identifying one or more matching images comprises:

for each obtained color image,

identifying one or more colors contained in the obtained color image;

for each identified color of the one or more identified colors contained in the obtained color image, comparing the identified color to the one or more search colors; and

designating the obtained color image a matching image based at least in part on a determination that one or more of the identified colors contained in the color image lies within at least one search color range.

4. The computer-implemented of claim 1, wherein obtaining the reference color palette comprises receiving the reference color palette from a user.

5. The computer-implemented method of claim 1, wherein obtaining the reference color palette comprises determining a color palette from an input image.

6. A computer-implemented method comprising:

under control of a hardware computing device configured with specific computer-executable instructions,

determining one or more search colors based at least in part on a reference color palette;

identifying one or more corresponding images from a plurality of color images, each corresponding image containing at least one color that corresponds to at least one of the one or more search colors; and

generating a grouping of items based at least in part on metadata associated with each corresponding image of the one or more corresponding images.

7. The computer-implemented method of claim 6 further comprising:

obtaining a color name; and

parsing the color name to determine the reference color palette.

8. The computer-implemented method of claim 6 further comprising determining a search color range for each search color of the one or more search colors, the search color range for each search color based at least in part on a color distance from the search color in a color space.

9. The computer-implemented method of claim 8, wherein the search color range for each search color is based at least in part on a human color distance formula.

10. The computer-implemented method of claim 6 further comprising:

obtaining a text query;

parsing the text query to identify one or more item identifiers; and

filtering the grouping of items to include items associated with at least one of the one or more item identifiers.

11. The computer-implemented method of claim 6 further comprising ordering the grouping of items based at least in part on a quantity of one or more search colors contained in the associated image.

12. A system comprising:

a data store configured to at least store computer-executable instructions; and

a hardware processor in communication with the data store, the hardware processor configured to execute the computer-executable instructions to at least;

determine one or more search colors based at least in part on a reference color palette;

identify one or more corresponding images from a plurality of color images, each corresponding image containing at least one color that corresponds to at least one of the one or more search colors; and

generate a grouping of items based at least in part on metadata associated with each corresponding image of the one or more corresponding images.

13. The system of claim 12 further comprising an item data store in communication with the hardware processor and configured to at least store the plurality of color images, wherein at least one color image of the plurality of color images is associated with an item and metadata.

14. The system of claim 12, wherein the reference color palette comprises a color palette name provided by a user.

15. The system of claim 12, wherein the reference color palette comprises a plurality of colors extracted from an input color image.

16. The system of claim 12, wherein the hardware processor is further configured to execute computer-executable instructions to at least generate a user interface, wherein the user interface is configured to display at least one of the one or more of corresponding images.

17. A non-transitory computer readable storage medium storing computer executable instructions that when executed by a processor perform operations comprising:

determining one or more search colors based at least in part on a reference color palette;

identifying one or more matching images from a plurality of color images, each matching image containing at least one color that corresponds to at least one of the one or more search colors; and

generating a grouping of items based at least in part on metadata associated with each matching image of the one or more matching images.

18. The non-transitory computer readable storage medium of claim 17, the operations further comprising determining a search color range for each search color of the one or more search colors, the search color range for each search color based at least in part on a color distance from the search color in a color space.

19. The non-transitory computer readable storage medium of claim 18, wherein identifying one or more matching images comprises:

for each color image of the plurality of color images,

identifying one or more colors contained in the color image;

for each identified color of the one or more identified colors contained in the color image, comparing the identified color to the one or more search colors; and

designating the color image a matching image based at least in part on a determination that one or more of the identified colors contained in the color image lies within at least one search color range.

20. The non-transitory computer readable storage medium of claim 18, wherein the search color range for each search color is based at least in part on a human color distance formula.