Smart categorization of artwork

• US 10,290,036 B1
• Filed: 12/04/2013
• Issued: 05/14/2019
• Est. Priority Date: 12/04/2013
• Status: Expired due to Fees

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Gella, Ganesh Kumar, Pitstick, Christopher, Shah, Pratik Prakash, Conduroglu, Atacan, Kirschner, Daniel Paul, Cherry, Jordan Carl
• Primary Examiner(s): Dunham, Jason B
• Assistant Examiner(s): Bargeon, Brittany E
• Application Number:
  US14/097,104
• Time in Patent Office: 1,987 Days
• Field of Search: 705 265
• CPC Class Codes: G06Q 30/0621 Item configuration or customization
• CPC Class Codes: G06Q 30/0623 Item investigation

Abstract

Techniques for obtaining information about an artwork and presenting such information may be provided. For example, a visual representation of the artwork, such as an image, may be received and analyzed. Results of the analysis may be categorized into a type and attributes of the artwork and may be presented to a first user for confirmation. Once a confirmation is received, the categorized type and attributes and other information may be presented to a second user searching for the artwork.

Claims

1. A computer-implemented method, comprising:

receiving, by a computer system based at least in part on an interaction between a computing service of a web site hosted on the computer system and a computing device of a provider of artwork, an image of the artwork from the computing device of the provider of the artwork;

in response to receiving the image, the computer system automatically at least;

determining a type of the artwork based at least in part on a comparison of the image to images of artworks stored in a database, the database associated with the web site;

determining, from the database, attributes of the artwork based at least in part on the type of the artwork, the database correlating the attributes to the type of the artwork, the attributes comprising a color attribute;

selecting an image analysis process based at least in part on the type and the attributes;

determining a first color for the color attribute based at least in part on an application of the image analysis process to the image, wherein the first color is determined by at least;

determining a color value of a pixel from the image,

mapping the color value to a color model to generate a mapped color,

adding the mapped color to a color group that corresponds to a color range, the color range (i) defined based at least in part on the color model and (ii) representing a color,

determining a distribution of the color group in the image based at least in part on mapped colors of the color group, and

setting the first color as the color represented by the color range based at least in part on a comparison of the distribution to a threshold;

determining an object in the artwork and a shape of the object based at least in part on the application of the image analysis process to the image, wherein the object and the shape are determined by at least detecting edges of the object and applying curve tracing to the detected edges;

setting a theme for the artwork to a predefined theme available from the database and associated with another artwork, the theme set based at least in part on matches between;

the first color and a predefined color of the other artwork, the object and a predefined object of the other artwork, and the shape of the object and a predefined shape of the predefined object; and

providing a first user interface to the computing device of the provider of the artwork, the first user interface provided in a first web page of the web site and presenting a text description of the first color, the object and the theme;

updating, by the computer system, the database by at least storing the first color, the object, the shape of the object, and the theme based at least in part on input received on the first user interface from the computing device of the provider; and

in response to a theme-based search for art from a user device received by a search engine of the web site, providing, by the computer system to the user device, a second user interface in a second web page of the web site, the second user interface presenting the type, the theme, and the image based at least in part on a match by the search engine between the theme-based search and the theme stored in the database.

2. The computer-implemented method of claim 1, wherein the type of the artwork comprises a multi-dimensional artwork.

3. The computer-implemented method of claim 1, wherein the artwork comprises a painting, and wherein the attributes comprise an orientation, a shape, a color palette, a subject matter, a style, an artist, a frame condition, or a creation date of the painting.

4. The computer-implemented method of claim 1, wherein the artwork comprises a sculpture, and wherein the attributes comprise an installation type, a shape, a color palette, a subject matter, a style, an artist, a material, or a creation date of the sculpture.

5. The computer-implemented method of claim 1, further receiving confirmation, from the computing device of the provider of the artwork, of at least one of the attributes, the confirmation received via the first user interface based at least in part on a confirmation option available on the first user interface.

6. The computer-implemented method of claim 5, wherein receiving confirmation of at least one of the attributes comprises:

receiving information from the computing device of the provider of the artwork, wherein the received information includes a description of another attribute that is not determined by the computer system; and

updating, by the computer system, at least one of;

a second color or the other attribute based at least in part on the received information.

7. The computer-implemented method of claim 1, further comprising:

wherein the first user interface presents a first field corresponding to the color attribute and populated with the first color, a second field populated with a second color, and a confirmation option selectable by the provider to confirm the second color.

8. The computer-implemented method of claim 1, wherein the image analysis process further comprises a machine learning algorithm trained based at least in part on a plurality of images of artworks.

9. The computer-implemented method of claim 1, wherein the artwork is a color painting or a poster, wherein the image analysis process comprises at least one of:

a machine learning algorithm, an edge detection algorithm, or a color distribution algorithm based at least in part on the type of the artwork corresponding to the color painting, and

a machine learning algorithm, or a facial recognition algorithm based at least in part on the type of the artwork corresponding to the poster.

10. The computer-implemented method of claim 1, further comprising:

determining, by the computer system, that a second color of the artwork is invalid, the second color determined based at least in part on the application of the image analysis process to the image, wherein the first user interface requests an update to the second color based at least in part on the second color being invalid; and

updating, by the computer system, the database to store the update to the second color, the object, and the shape of the object based at least in part on the input received on the first user interface from the computing device of the provider.

11. A computer-implemented method, comprising:

receiving, by a computer system configured with executable instructions associated with a network-based resource, an image of a piece of art;

in response to receiving the image, the computer system automatically at least;

determining a type of the piece of art based at least in part on a comparison of the image to images of pieces of art stored in a database, the database associated with the network-based resource;

determining, from the database, attributes of the piece of art based at least in part on the type of the piece of art, the database correlating the attributes with the type of the piece of art, the attributes comprising a color attribute;

selecting an image analysis process based at least in part on the type of the piece of art and the attributes;

determining a first color for the color attribute based at least in part on an application of the image analysis process to the image, wherein the first color is determined by at least;

determining a color value of a pixel from the image,

mapping the color value to a color model to generate a mapped color,

adding the mapped color to a color group that corresponds to a color range, the color range (i) defined based at least in part on the color model and (ii) representing a color,

determining a distribution of the color group in the image based at least in part on mapped colors of the color group, and

setting the first color as the color represented by the range based at least in part on a comparison of the color distribution to a threshold;

determining an object in the piece of art and a shape of the object based at least in part on the application of the image analysis process to the image, wherein the object and the shape are determined by at least detecting edges of the object and applying curve tracing to the detected edges; and

setting a theme for the piece of art to a predefined theme available from the database and associated with another piece of art, the theme set based at least in part on matches between;

the first color and a predefined color of the other piece of art, the object and a predefined object of the other piece of art, and the shape of the object and a predefined shape of the predefined object; and

providing, by the computer system in a network document of the network-based resource, information about the piece of art to a computing device of a user of an electronic marketplace, the information comprising the type, the first color, the theme, and the image of the piece of art.

12. The computer-implemented method of claim 11, wherein the image is received from the computing device.

13. The computer-implemented method of claim 11, further comprising:

receiving a description of the piece of art from a provider of the piece of art; and

updating an attribute of the piece of art based at least in part on the received description.

14. The computer-implemented method of claim 11, further comprising:

instructing the computing device to present a first indication that the first color is valid.

15. The computer-implemented method of claim 14, wherein the first indication is indicative that the computer system does not request input from the user about the first color.

16. The computer-implemented method of claim 11, wherein a second color of the piece of art is determined based at least in part on the theme.

17. The computer-implemented method of claim 16, wherein the attributes further comprise another attribute, and wherein the other attribute is determined based at least in part on the first color and the type of the piece of art.

18. A system, comprising:

at least one memory that stores computer-executable instructions associated with a network-based resource; and

at least one processor configured to access the at least one memory, wherein the at least one processor is configured to execute the computer-executable instructions to collectively at least;

receive information about an artwork, the information comprising a visual representation of the artwork;

in response to receiving the information, automatically;

determine, based at least in part on a comparison of the visual representation to another visual representation stored in a database, a type of the artwork, the database associated with the network-based resource;

determine, from the database and based at least in part on the type, a list of attributes of the artwork from the database, the database correlating the attributes with the type of the artwork, the list comprising a color attribute;

select an image analysis process based at least in part on the type of the artwork and the list of attributes;

determine a first color for the color attribute based at least in part on an application of the image analysis process to the visual representation wherein the first color is determined by at least;

determining a color value of a pixel from the visual representation,

mapping the color value to a color model to generate a mapped color,

adding the mapped color to a color group that corresponds to a color range, the color range (i) defined based at least in part on the color model and (ii) representing a color,

determining a distribution of the color group in the visual representation based at least in part on mapped colors of the color group, and

setting the first color as the color represented by the range based at least in part on a comparison of the distribution to a threshold;

determine an object in the artwork and a shape of the object based at least in part on the application of the image analysis process to the visual representation, wherein the object and the shape are determined by at least detecting edges of the object and applying curve tracing to the detected edges; and

set a theme for the artwork to a predefined theme available from the database and associated with another artwork, the theme set based at least in part on matches between;

the first color and a predefined color of the other artwork, the object and a predefined object of the other artwork, and the shape of the object and a predefined shape of the predefined object; and

provide, in a network document of the network-based resource, information that comprises the visual representation, the first color, and the theme to a computing device of a user of an electronic marketplace.

19. The system of claim 18, wherein the visual representation of the artwork comprises an image of the artwork, and wherein the type is determined based at least in part on the image.

20. The system of claim 18, wherein the computer-readable instructions further comprise instructions that, when executed by the at least one processor, cause the system to at least:

determine an attribute from the list of attributes that cannot be extracted from the visual representation; and

send a request to a user device of another user to select the attribute.

21. The system of claim 18, wherein the computer-readable instructions further comprise instructions that, when executed by the at least one processor, cause the system to at least:

receive, from the user, information indicative of a confirmation of the theme; and

in response to a second user searching for art, present a search criteria based at least in part on the confirmation of the theme.

22. One or more non-transitory computer-readable storage media storing computer-executable instructions that, when executed by one or more computer systems, configure the one or more computer systems to perform operations comprising:

receiving information associated with an artwork, the information comprising a visual representation of the artwork;

in response to receiving the information, automatically;

determining, based at least in part on the received information, a type of the artwork, the type being indicative of a color attribute of the artwork and determined based at least in part on a comparison of the visual representation with another visual representation stored in a database associated with a network-based resource, the database correlating attributes with the type of the piece of artwork;

selecting an image analysis process based at least in part on the type of the artwork and the color attribute;

determining, based at least in part on an application of the image analysis process to the visual representation of the artwork, first color for the color attribute, wherein the first color is determined by at least;

determining a color value of a pixel from the visual representation,

mapping the color value to a color model to generate a mapped color,

adding the mapped color to a color group that corresponds to a color range, the color range (i) defined based at least in part on the color model and (ii) representing a color,

determining a distribution of the color group in the visual representation based at least in part on mapped colors of the color group, and

setting the first color as the color represented by the range based at least in part on a comparison of the distribution to a threshold;

determining an object in the artwork and a shape of the object based at least in part on the application of the image analysis process to the visual representation, wherein the object and the shape are determined by at least detecting edges of the object and applying curve tracing to the detected edges; and

setting a theme for the artwork to a predefined theme available from the database and associated with another artwork, the theme set based at least in part on matches between;

the first color and a predefined color of the other artwork, the object and a predefined object of the other artwork, and the shape of the object and a predefined shape of the predefined object; and

providing, in a network document of the network-based resource to a computing device of a user of an electronic marketplace, information about the type, the color attribute, and the theme of the artwork.

23. The one or more non-transitory computer-readable storage media of claim 22, wherein the providing comprises instructing the computing device of the user to present the information to the user.

24. The one or more non-transitory computer-readable storage media of claim 23, wherein the computer-readable instructions, when executed by the one or more computer systems, further configure the one or more computer systems to perform operations comprising presenting an interface at the computing device, wherein the interface allows the user to search for the artwork based at least in part on at least one of the type of the artwork, the color attribute of the artwork, or the theme of the artwork.