System and method for providing an interactive shopping experience via webcam

• US 9,652,654 B2
• Filed: 06/04/2012
• Issued: 05/16/2017
• Est. Priority Date: 06/04/2012
• Status: Active Grant

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Li, Bin, Yun, Xiaohan, Tan, Zehong
• Primary Examiner(s): Park, Chan
• Assistant Examiner(s): SCHWARTZ, RAPHAEL M
• Application Number:
  US13/488,020
• Publication Number:
  US 20130322685A1
• Time in Patent Office: 1,807 Days
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06Q 30/0241 Advertisements
• CPC Class Codes: G06Q 30/0276 Advertisement creation
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0641 Shopping interfaces
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: G06T 11/00 2D [Two Dimensional] image generation
• CPC Class Codes: G06V 40/165 using facial parts and geometric relationships
• CPC Class Codes: H04N 5/265 Mixing

Abstract

A system and method for providing an interactive shopping experience via webcam is disclosed. A particular embodiment includes enabling a user to select from a plurality of items of virtual apparel; obtaining an image of a user via a web-enabled camera (webcam); using a data processor to perform facial detection on the image to isolate an image of a face of the user; estimating the user'"'"'s position according to a position and a size of the image of the user'"'"'s face; modifying an image corresponding to the selected item of virtual apparel based on the size of the image of the user'"'"'s face; and auto-fitting the modified image corresponding to the selected item of virtual apparel to the image of the user'"'"'s face.

Claims

1. A method comprising:

displaying a plurality of items of virtual apparel for a user to select from, each of the items of virtual apparel comprising a pre-stored image of the respective item;

receiving a selection of one of the items of virtual apparel;

obtaining a first image of a user via a web-enabled camera (webcam);

performing, using a processor, facial detection, the performing comprising;

identifying an available set of classifiers, each classifier in the available set of classifiers including a feature that provides differentiation between a face image and a non-face image;

extracting a subset of classifiers from the available set of classifiers and assigning a specific weight to each classifier of the subset;

identifying a second image from the first image, the second image being a region of the first image and identified based on a score associated with the region of the first image transgressing a threshold score relative to a total weight of the subset of classifiers;

determining, based on the transgressing of the threshold score, that the second image includes a face; and

isolating the second image from the first image;

estimating, based on the isolating of the second image, a position of a body of the user relative to a position and a size of the second image, the body of the user and the estimated position of the body of the user being separate from and below the second image;

modifying a pre-stored image corresponding to the selected item of virtual apparel, the modifying being relative to a portion of the body of the user that is separate from and below the second image and based on the size of the second image; and

auto-fitting the modified image corresponding to the selected item of virtual apparel to the estimated position of the body of the user to appear separate from and below the second image.

2. The method as claimed in claim 1 wherein the facial detection includes searching the first image of the user in a plurality of pixel regions.

3. The method as claimed in claim 1, further comprising:

performing skin detection and skin color tracking.

4. The method as claimed in claim 1, further comprising:

converting RGB pixels to YCbCr color space.

5. The method as claimed in claim 1, further comprising:

providing a virtual button click for the user to modify the auto-fitted image corresponding to the selected item of virtual apparel.

6. The method as claimed in claim 1, further comprising:

providing gesture recognition, the gesture recognition including a gesture move function.

7. The method as claimed in claim 1, further comprising:

providing gesture recognition, the gesture recognition including a gesture stretch horizontally function.

8. The method as claimed in claim 1, further comprising:

providing gesture recognition, the gesture recognition including a gesture stretch diagonally function.

9. A system comprising:

one or more processors and executable instructions accessible on a computer-readable medium that, when executed, cause the one or more processors to perform operations comprising;

displaying a plurality of items of virtual apparel for a user to select from, each of the items of virtual apparel comprising a pre-stored image of the respective item;

receiving a selection of one of the items of virtual apparel;

obtaining a first image of a user via a web-enabled camera (webcam);

performing facial detection, the performing comprising;

identifying an available set of classifiers, each classifier in the available set of classifiers including a feature that provides differentiation between a face image and a non-face image;

extracting a subset of classifiers from the available set of classifiers and assigning a specific weight to each classifier of the subset;

identifying a second image from the first image, the second image being a region of the first image and identified based on a score associated with the region of the first image transgressing a threshold score relative to a total weight of the subset of classifiers;

determining, based on the transgressing of the threshold score, that the second image includes a face; and

isolating the second image from the first image;

estimating, based on the isolating of the second image, a position of a body of the user relative to a position and a size of the second image, the body of the user and the estimated position of the body of the user being separate from and below the second image;

modifying a pre-stored image corresponding to the selected item of virtual apparel, the modifying being relative to a portion of the body of the user that is separate from and below the second image and based on the size of the second image; and

auto-fitting the modified image corresponding to the selected item of virtual apparel to the estimated position of the body of the user to appear separate from and below the second image.

10. The system as claimed in claim 9 wherein the performing the facial detection further comprises searching the first image of the user in a plurality of pixel regions.

11. The system as claimed in claim 9, wherein the operations further comprise performing skin detection and skin color tracking.

12. The system as claimed in claim 9, wherein the operations further comprise converting RGB pixels to YCbCr color space.

13. The system as claimed in claim 9, wherein the operations further comprise providing a virtual button click for the user to modify the auto-fitted image corresponding to the selected item of virtual apparel.

14. The system as claimed in claim 9, wherein the operations further comprise providing gesture recognition, the gesture recognition including a gesture stretch horizontally function.

15. The system as claimed in claim 9, wherein the operations further comprise providing gesture recognition, the gesture recognition including a gesture stretch diagonally function.

17. The system as claimed in claim 9 wherein the at least one processor is further configured to provide gesture recognition, the gesture recognition including a gesture grab and move function to grab and move the selected item of virtual apparel.

16. A hardware storage device having stored therein instructions which, when executed by one or more processors of a machine, cause the machine to perform operations comprising:

displaying a plurality of items of virtual apparel for a user to select from, each of the items of virtual apparel comprising a pre-stored image of the respective item;

receiving a selection of one of the items of virtual apparel;

obtaining a first image of a user via a web-enabled camera (webcam);

performing facial detection, the performing comprising;

identifying an available set of classifiers, each classifier in the available set of classifiers including a feature that provides differentiation between a face image and a non-face image;

extracting a subset of classifiers from the available set of classifiers and assigning a specific weight to each classifier of the subset;

identifying a second image from the first image, the second image being a region of the first image and identified based on a score associated with the region of the first image transgressing a threshold score relative to a total weight of the subset of classifiers;

determining, based on the transgressing of the threshold score, that the second image includes a face; and

isolating the second image from the first image;

estimating, based on the isolating of the second image, a position of a body of the user relative to a position and a size of the second image, the body of the user and the estimated position of the body of the user being separate from and below the second image;

modifying a pre-stored image corresponding to the selected item of virtual apparel, the modifying being relative to a portion of the body of the user that is separate from and below the second image and based on the size of the second image; and

auto-fitting the modified image corresponding to the selected item of virtual apparel to the estimated position of the body of the user to appear separate from and below the second image.