US 12,033,436 B2
Methods and apparatus for human pose estimation from images using dynamic multi-headed convolutional attention
Alec Diaz-Arias, Columbia, MO (US); Dmitriy Shin, Columbia, MO (US); Jean E. Robillard, Iowa City, IA (US); Mitchell Messmore, Iowa City, IA (US); and John Rachid, Iowa City, IA (US)
Assigned to INSEER Inc., Iowa City, IA (US)
Filed by INSEER Inc., Iowa City, IA (US)
Filed on Sep. 14, 2022, as Appl. No. 17/944,418.
Application 17/944,418 is a continuation of application No. 17/740,650, filed on May 10, 2022, granted, now 11,482,048.
Prior Publication US 2023/0368578 A1, Nov. 16, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06K 9/00 (2022.01); G06V 10/80 (2022.01); G06V 40/10 (2022.01); G06V 40/20 (2022.01); G16H 20/30 (2018.01)
CPC G06V 40/23 (2022.01) [G06V 10/806 (2022.01); G06V 40/103 (2022.01); G16H 20/30 (2018.01)] 16 Claims
OG exemplary drawing
 
1. An apparatus, comprising:
at least a processor; and
a memory operatively coupled to the processor, the memory storing instructions that when executed cause the processor to:
receive a plurality of image frames, each image frame from the plurality of image frames containing a measured temporal joint data of a subject;
receive a plurality of joint localization overlays from a spatial joint machine-learning model;
execute a trained limb segment machine-learning model to identify a plurality of frame interrelations using the plurality of image frames and the plurality of joint localization overlays as an input, the trained limb segment machine-learning model trained using an interrelation training set that contains limb segment image data correlated to a limb matrix in a motion sequence; and
generate a temporal joints profile based on the plurality of frame interrelation.