US 11,531,697 B2 · Identifying and providing digital images depicting human poses utilizing visual interactive content search and virtual mannequins

Abstract

The present disclosure relates to systems, methods, and non-transitory computer readable media for accurately, efficiently, and flexibly identifying and providing digital images of human figures in poses corresponding to a query pose. In particular, the disclosed systems can provide multiple approaches to searching for and providing pose images, including identifying a digital image depicting a human figure in a particular pose based on a query digital image that depicts the pose or identifying a digital image depicting a human figure in a particular pose based on a virtual mannequin. Indeed, the disclosed systems can provide a manipulable virtual mannequin that defines a query pose for searching a repository of digital images. Additionally, the disclosed systems can generate and provide digital pose image groups by clustering digital images together according to poses of human figures within a pose feature space.

Claims20

Dependent claims

A computer-implemented method for generating and providing pose-based search interfaces, the computer-implemented method comprising:

providing, for display within a digital image search interface, a keyword query element for entering a keyword query and a digital image element for selecting a query image depicting a human figure;
generating a query pose from the human figure of the query image utilizing a pose neural network;
determining, utilizing the pose neural network, a plurality of candidate digital poses by processing a plurality of digital images from a repository of digital images that include digital content reflecting the keyword query;
generating, by comparing the query pose from the query image and candidate digital poses from the plurality of digital images, a query response comprising a digital image from the plurality of digital images that depicts digital content reflecting the keyword query and that further depicts a human figure with a candidate digital pose corresponding to the query pose;
and
providing the query response for display within the digital image search interface together with the keyword query element and the digital image element.

3 more elements

A system comprising:

one or more memory devices comprising a repository of digital images and a pose neural network;
and
one or more computing devices configured to cause the system to:
in response to user selection of a digital image depicting a human figure, determine a query pose of the human figure by processing the digital image utilizing the pose neural network;
utilize the query pose to generate, for the digital image depicting the human figure, a virtual mannequin comprising manipulable joints and segments arranged according to the query pose;
based on user interaction modifying one or more manipulable joints or segments of the virtual mannequin, generate a modified query pose reflecting modifications to the virtual mannequin;
identify, from the repository of digital images, digital images depicting human figures with candidate digital poses corresponding to the modified query pose of the virtual mannequin;
and
provide, for display within a user interface via a client device, the virtual mannequin and the digital images depicting human figures with candidate digital poses corresponding to the modified query pose.

6 more elements

A non-transitory computer readable medium comprising instructions that, when executed by at least one processor, cause a computing device to:

in response to receiving a keyword query, generate a plurality of digital pose image groups by:
identifying, from a repository of digital images, a plurality of digital images portraying digital content corresponding to the keyword query and depicting human figures;
determining poses for the plurality of digital images depicting human figures from the repository of digital images;
and
clustering the plurality of digital images into digital pose image groups by comparing pose vectors of the determined poses of the human figures;
and
provide, for display via a client device, selectable elements corresponding to the digital pose image groups and the keyword query.

4 more elements

Dependencies are inferred from the claim text.

Identifying and providing digital images depicting human poses utilizing visual interactive content search and virtual mannequins

Life of this patent

Identifying and providing digital images depicting human poses utilizing visual interactive content search and virtual mannequins

Life of this patent

Abstract

Claims20