Pose recommendation and real-time guidance for user-generated content

Patent metadata

Field	Value
Publication number	US-12094032-B2
Application number	US-202217946202-A
Country	US
Kind code	B2
Filing date	Sep 16, 2022
Priority date	Sep 16, 2022
Publication date	Sep 17, 2024
Grant date	Sep 17, 2024

Abstract

In some embodiments, techniques for producing user-generated content are provided. For example, a process may involve sending a product identifier; receiving a first candidate image that is associated with the product identifier; determining that a similarity between a user structure and a target structure satisfies a threshold condition, wherein the user structure characterizes a figure of a user in a first input image and the target structure is based on a pose guide associated with the first candidate image; and capturing, based on the determining, the first input image.

First claim

The invention claimed is:

1. A method comprising:
sending, by a pose-recommending module, a product identifier;
receiving, by the pose-recommending module, a first candidate image that is associated with the product identifier, wherein the product identifier is not derived by the pose-recommending module;
determining, by a similarity-determining module, that a similarity between a user structure and a target structure satisfies a threshold condition, wherein the user structure characterizes a figure of a user in a first input image and the target structure is based on a pose guide associated with the first candidate image; and
capturing, by an image-capturing module and based on the determination, the first input image.

2. The method of claim 1 , wherein the product identifier indicates a name of a product, a category of the product, a universal product code (UPC), or a stock keeping unit (SKU).

3. The method of claim 1 , wherein the sending comprises sending, by the pose-recommending module, a request that includes the product identifier and also includes contextual information obtained from a user profile, from a user selection, or from at least one image provided by a camera interface.

4. The method of claim 1 , wherein the method comprises:
receiving, by the pose-recommending module, a plurality of candidate images that are associated with the product identifier, wherein the plurality of candidate images includes the first candidate image; and
receiving, by the pose-recommending module, an indication that the user has selected the first candidate image.

5. The method of claim 4 , wherein the method further comprises displaying, on a display of a computing device, the plurality of candidate images.

6. The method of claim 1 , wherein the pose guide describes a first structure that characterizes a human figure in the first candidate image.

7. The method of claim 6 , wherein the user structure comprises at least one of a contour of the figure of the user and a skeleton of the figure of the user; and
wherein the first structure comprises at least one of a contour of the human figure and a skeleton of the human figure.

8. The method of claim 6 , wherein the method comprises producing the target structure, by a motion retargeting module, by posing the user structure in a pose described by the pose guide.

9. The method of claim 1 , wherein the method further comprises:
receiving, by an image analysis module, a time-ordered plurality of input images of the user, wherein the time-ordered plurality of input images includes the first input image; and
obtaining, by the image analysis module, the user structure from at least the first input image.

10. The method of claim 9 , wherein the method further comprises displaying an augmented image in which a second target structure that is based on the pose guide is overlaid on a second input image of the time-ordered plurality of input images.

11. The method of claim 1 , wherein the similarity between the user structure and the target structure is based on distances between corresponding features of the target structure and the user structure.

12. The method of claim 1 , wherein the method further comprises sending, by the pose-recommending module and based on the determination, an indication of use of the first candidate image.

13. The method of claim 1 , wherein the capturing comprises at least one of: copying the first input image to a buffer, or preventing a buffer in which the first input image is stored from being overwritten.

14. A non-transitory computer-readable medium having program code that is stored thereon, the program code executable by one or more processing devices for performing operations comprising:
sending a product identifier;
receiving a plurality of candidate images that are associated with the product identifier, wherein the product identifier is not derived by a pose-recommending module;
receiving an indication that a user has selected a first candidate image from among the plurality of candidate images;
determining that a similarity between a user structure and a target structure satisfies a threshold condition, wherein the user structure characterizes a figure of the user in a first input image and the target structure is based on a pose guide associated with the first candidate image; and
capturing, based on the determining, the first input image.

15. The non-transitory computer-readable medium of claim 14 , wherein the product identifier indicates a name of a product, a category of the product, a universal product code (UPC), or a stock keeping unit (SKU).

16. The non-transitory computer-readable medium of claim 14 , wherein:
the pose guide describes a first structure that characterizes a human figure in the first candidate image;
the first structure comprises a skeleton of the human figure;
the user structure comprises a skeleton of the figure of the user; and
the operations further comprise producing the target structure, by a motion retargeting module, by posing the user structure in a pose described by the pose guide.

17. The non-transitory computer-readable medium of claim 14 , wherein the operations further comprise:
receiving a time-ordered plurality of input images of the user, wherein the time-ordered plurality of input images includes the first input image;
obtaining the user structure from at least the first input image; and
displaying an augmented image in which a second target structure that is based on the pose guide is overlaid on a second input image of the time-ordered plurality of input images.

18. The non-transitory computer-readable medium of claim 14 , wherein the similarity between the user structure and the target structure is based on distances between corresponding features of the target structure and the user structure.

19. A computing device comprising:
a pose-recommending module configured to:
send a request that includes a product identifier;
receive a plurality of candidate images that are associated with the product identifier, wherein the product identifier is not derived by the pose-recommending module; and
receive an indication that a user has selected a first candidate image from among the plurality of candidate images;

a similarity-determining module configured to determine that a similarity between a user structure and a target structure satisfies a threshold condition, wherein the user structure characterizes a figure of the user in a first input image and the target structure is based on a pose guide associated with the first candidate image; and
an image-capturing module configured to capture, based on the determination, the first input image.

20. The computing device of claim 19 , wherein:
the pose guide describes a first structure that characterizes a human figure in the first candidate image;
the first structure comprises a skeleton of the human figure;
the user structure comprises a skeleton of the figure of the user; and
the computing device further comprises a motion retargeting module configured to produce the target structure by posing the user structure in a pose described by the pose guide.

Assignees

• Adobe Inc

Inventors

• Singhal Gourav
• Das Tridib
• Gupta Sourabh

Classifications

• H04N23/611

  where the recognised objects include parts of the human body · CPC title

• G06T7/251

  involving models · CPC title

• G06Q30/0627

  by specifying product or service characteristics, e.g. product dimensions · CPC title

• G06Q30/0643

  graphically representing goods, e.g. 3D product representation · CPC title

• G06V40/28

  Recognition of hand or arm movements, e.g. recognition of deaf sign language (static hand signs G06V40/113) · CPC title