US 12,322,210 B2
Computer-aided identification of ergonomic risk
Stefano Celestini, Oakville (CA)
Assigned to PACEFACTORY INC., Oakville (CA)
Filed by Pacefactory Inc., Oakville (CA)
Filed on Feb. 2, 2023, as Appl. No. 18/105,151.
Prior Publication US 2023/0177883 A1, Jun. 8, 2023
Int. Cl. G06K 9/00 (2022.01); G06V 20/40 (2022.01); G06V 40/20 (2022.01)
CPC G06V 40/23 (2022.01) [G06V 20/41 (2022.01)] 21 Claims
OG exemplary drawing
 
1. A computer-implemented method for generating ergonomics-annotated human forms and location histories in a video sequence, the method comprising:
receiving a video sequence comprising a plurality of images representing a scene;
analyzing the images to identify at least one human form of a respective human within the scene;
generating a body wireframe for each human form within the scene in each one of the images;
analyzing each body wireframe for each image to identify at least one ergonomic risk factor;
tracking movement of each human form within the scene;
for each human form within the scene, superimposing a respective footprint on each of the images, wherein the respective footprints combine to represent a location history of the respective human form within the workspace in the video sequence;
for each ergonomic risk factor, annotating the video sequence with the respective ergonomic risk factor at time of occurrence within the video sequence, wherein:
a first annotation for the respective ergonomic risk factor is associated with the respective human form in the video sequence; and
a second annotation for the respective ergonomic risk factor is associated with the respective location history in the video sequence.