Interactive interfaces for generating annotation information

• US 10,379,721 B1
• Filed: 11/28/2016
• Issued: 08/13/2019
• Est. Priority Date: 11/28/2016
• Status: Active Grant

• Current Assignee: A9.com Incorporated (Amazon.com, Inc.)
• Original Assignee: A9.com Incorporated (Amazon.com, Inc.)
• Inventors: Natesh, Aishwarya, Zamalieva, Daniya
• Primary Examiner(s): Motsinger, Sean T
• Application Number:
  US15/362,600
• Time in Patent Office: 988 Days
• CPC Class Codes: G06F 3/0482 Interaction with lists of selectable items, e.g. menus
• CPC Class Codes: G06F 3/04845 for image manipulation, e.g. dragging, rotation, expansion or change of colour
• CPC Class Codes: G06Q 30/0641 Shopping interfaces
• CPC Class Codes: G06T 2200/24 involving graphical user interfaces [GUIs]
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/90 Determination of colour characteristics

Abstract

Approaches provide for an interactive interface for determining annotation information that can be used to recommend content, improve user interaction and satisfaction, as well as improve various image recognition approaches. For example, a user can be presented an interactive interface, such as a game. The interactive interface can include colored graphical elements, where each graphical element can be associated with and display a visual feature of an item. A user can be provided a task or goal, such as to select, move, or otherwise arrange graphical elements based on a visual feature of the elements. The user interaction can generate annotation information that can be utilized to determine related content as well as improve various image recognition approaches.

Claims

1. A computing device comprising:

at least one processor;

a camera; and

memory storing instructions that, when executed by the at least one processor, cause the computing device to perform operations, comprising;

obtain image data that includes a representation of an object using the camera;

analyze the image data to attempt to recognize the object;

determine an image recognition confidence score below a threshold confidence score, the image recognition confidence score indicating a likelihood of recognizing the object;

segment the image data into a plurality of regions;

analyze the plurality of regions to determine a color descriptor for each of the plurality of regions, the color descriptor for each of the plurality of regions describing a color represented in image data for a respective region;

compare respective color descriptors against a stored set of color descriptors to determine a plurality of visual similarity scores, each of the stored set of color descriptors associated with a patch of a product that is associated with color information;

determine a set of patches based at least in part on the plurality of visual similarity scores;

generate an interface that includes a plurality of graphical elements, each graphical element associated with a patch of the set of patches and displaying a color of its associated patch;

receive a user selection of adjacent graphical elements; and

generate color annotation information based at least in part on the adjacent graphical elements.

2. The computing device of claim 1, wherein the instructions when executed to determine the set of patches further enable the computing device to:

identify visual similarity scores from the plurality of visual similarity scores that satisfy a selection criterion.

3. The computing device of claim 1, wherein the instructions when executed to analyze the plurality of regions to determine the color descriptor for each of the plurality of regions further enable the computing device to:

use a trained classify to determine the color descriptor for each of the plurality of regions.

4. The computing device of claim 1, wherein the instructions when executed to determine the set of patches further enable the computing device to:

receive a selection of a product;

display a product page associated with the product, the product being associated with the annotation information; and

display on the product page with the product a listing of one or more products that are associated with the product based at least in part on the annotation information.

5. A computer-implemented method, comprising:

obtaining image data that includes a representation of an object;

determining a visual descriptor for an image region of the image data;

associating the visual descriptor with a patch of color of a plurality of patches of color based at least in part on a similarity between the visual descriptor and the patch of color, each of the patches of color associated with an area of a product of an electronic catalog of products;

generating an interface that includes a plurality of graphical elements, one of the plurality of graphical elements associated with the patch of color, at least a portion of the remaining plurality of graphical elements associated with one of the plurality of patches of color and displaying a color of its associated patch of color; and

in response to receiving a user interaction of the graphical elements, generating annotation information.

6. The computer-implemented method of claim 5, further comprising:

analyzing the image data using a recognition algorithm to attempt to recognize the object;

determining an image recognition confidence score based at least in part on the analyzing;

comparing the image recognition confidence score to a threshold confidence score; and

determining that the recognition algorithm failed.

7. The computer-implemented method of claim 5, further comprising:

comparing the visual descriptor against a stored visual descriptor associated with the patch of color to determine a visual similarity score; and

determining that the visual similarity score satisfies a selection criterion.

8. The computer-implemented method of claim 7, wherein the selection criterion includes at least one of identifying a highest value visual similarity score or a visual similarity score that satisfies a threshold value.

9. The computer-implemented method of claim 5, further comprising:

receiving a search query for a product;

receiving a selection of a color preference for the product; and

displaying a listing of products based at least in part on the search query and the color preference, and listing one product of the listing of products further based at least in part on the annotation information.

10. The computer-implemented method of claim 5, wherein the visual descriptor describes visual features of the image region based at least in part on patterns represented in the image region, texture represented in the image region, or color represented in the image region.

11. The computer-implemented method of claim 5, wherein the visual descriptor includes color information for a first color and a second color, the method comprising:

determining a first percentage of representation of the first color represented in the image region;

determining a second percentage of representation of the second color represented in the image region;

determining the first percentage of representation is greater than the second percentage of representation; and

associating the visual descriptor with the first color.

12. The computer-implemented method of claim 5, further comprising using a trained classifier to determine the visual descriptor for the image region.

13. The computer-implemented method of claim 5, further comprising:

displaying the interface in response to one of a failed visual search, at a predetermined interval of time, upon completion of an electronic checkout process, or in response to adding a product to a virtual shopping cart.

14. The computer-implemented method of claim 5, wherein the plurality of graphical elements are associated with one of a color attribute that specifies a color of the graphical element, a size attribute that specifies a size of the graphical element, an animation element that specifies animation characteristics of the graphical element.

15. The computer-implemented method of claim 5, wherein the user interaction includes at least one of selecting adjacent graphical elements, moving graphical elements, or selecting graphical elements.

16. A computing system, comprising:

at least one processor; and

memory including instructions that, when executed by the at least one processor, cause the computing system to;

obtain image data that includes a representation of an object;

determine a visual descriptor for an image region of the image data;

associate the visual descriptor with a patch of color of a plurality of patches of color based at least in part on a similarity between the visual descriptor and the patch of color, each of the patches of color associated with an area of a product of an electronic catalog of products;

generate an interface that includes a plurality of graphical elements, each element of the plurality of graphical elements associated with one of the plurality of patches of color and displaying a color of its associated patch of color; and

in response to receiving a selection of adjacent graphical elements, generate annotation information.

17. The computing system of claim 16, wherein the instructions when executed further enable the computing system to:

analyzing the image data using a recognition algorithm to attempt to recognize the object;

determine an image recognition confidence score based at least in part on the analyzing;

compare the image recognition confidence score to a threshold confidence score; and

determine that the recognition algorithm failed.

18. The computing system of claim 16, wherein the instructions when executed further enable the computing system to:

compare the visual descriptor against a stored visual descriptor associated with the patch of color to determine a visual similarity score; and

determine that the visual similarity score satisfies a selection criterion, wherein the selection criterion includes at least one of identifying a highest value visual similarity score or a visual similarity score that satisfies a threshold value.

19. The computing system of claim 16, wherein the instructions when executed further enable the computing system to:

receive a search query for a product;

receive a selection of a color preference for the product; and

display a listing of products based at least in part on the search query and the color preference, and listing one product of the listing of products further based at least in part on the annotation information.

20. The computing system of claim 16, wherein the instructions when executed further enable the computing system to:

display the interface in response to one of a failed visual search, at a predetermined interval of time, upon completion of an electronic checkout process, or in response to adding a product to a virtual shopping cart.