System and method of pose detection using a composite image
A method of pose detection includes the steps of receiving a first image of a subject in an original orientation, generating a second image of the subject rotated 90 degrees from the original orientation, a third image of the subject rotated 180 degrees from the original orientation, and a fourth image of the subject rotated 270 degrees from the original orientation, generating a composite image comprising the first, second, third and fourth image, detecting one or more poses in the composite image, selecting a detected pose of the subject from the composite image based on a usability criteria, rotating the composite image such that the detected pose is in the original orientation, and transmitting the detected pose in the original orientation to a remote server. A system for pose detection and alternate methods and systems for pose detection are also disclosed.
1 . A method of pose detection comprising:
receiving from a controller communicatively coupled to a camera a first image of a subject in an original orientation;
generating a second image of the subject rotated 90 degrees from the original orientation, a third image of the subject rotated 180 degrees from the original orientation, and a fourth image of the subject rotated 270 degrees from the original orientation;
generating a composite image comprising the first, second, third and fourth image;
detecting one or more poses in the composite image;
selecting a detected pose of the subject from the composite image based on a usability criteria;
rotating the composite image such that the detected pose is in the original orientation; and
transmitting the detected pose in the original orientation to a remote server;
wherein the composite image width equals the sum of the width and height of the first image; and
wherein selecting a detected pose of the subject comprises processing the composite image through a pose detection neural network.
2 . The method of claim 1 further comprising:
determining dimensions of the received image.
3 . The method of claim 2 , wherein the composite image height equals the sum of the width and height of the first image.
4 . The method of claim 2 , wherein the composite image comprises first, second, third and fourth image configured about its perimeter.
5 . The method of claim 1 further comprising:
checking for usability comprises detecting visibility of at least one side of the body.
6 . The method of claim 5 , wherein detecting visibility comprises detecting at least a shoulder, hip, and knee.
7 . A method for monitoring of a subject comprising:
the method of claim 1 , wherein the first image is generated by a camera located in a residential environment.
8 . A system for pose detection comprising:
a camera communicatively coupled to a controller, the controller configured to:
receive a first image of a subject in an original orientation;
generate a second image of the subject rotated 90 degrees from the original orientation, a third image of the subject rotated 180 degrees from the original orientation, and a fourth image of the subject rotated 270 degrees from the original orientation;
generate a composite image comprising the first, second, third and fourth image;
select an image of the subject from the composite image based on a usability criteria;
detect a pose in the selected image;
rotate the detected pose back to the original orientation; and
transmit the detected pose in the original orientation to a remote server;
wherein the composite image width equals the sum of the width and height of the first image; and
wherein selecting the image of the subject comprises processing the composite image through a pose detection neural network.
9 . The system of claim 8 , wherein the controller is configured to:
determine dimensions of the received image.
10 . The system of claim 9 , wherein the composite image height equals the sum of the width and height of the first image.
11 . The system of claim 9 , wherein the composite image comprises first, second, third and fourth image configured about its perimeter.
12 . The system of claim 9 , wherein checking for usability comprises detecting visibility of at least one side of the body.
13 . The system of claim 12 , wherein detecting visibility comprises detecting at least a shoulder, hip, and knee.
14 . A system for monitoring of a subject comprising:
the system of claim 8 , wherein the first image is generated by a camera located in a residential environment.