US 12,073,027 B2
Behavior-based standard operating procedure detection
Keyu Qi, Dongguan (CN); Hailing Zhou, Shenzhen (CN); Nan Ke, Xiamen (CN); David Nguyen, Newark, CA (US); and Binghao Tang, Shenzhen (CN)
Assigned to ACCENTURE GLOBAL SOLUTIONS LIMITED, Dublin (IE)
Filed by Accenture Global Solutions Limited, Dublin (IE)
Filed on Dec. 20, 2022, as Appl. No. 18/084,885.
Prior Publication US 2024/0201789 A1, Jun. 20, 2024
Int. Cl. G06F 3/01 (2006.01); G06V 10/74 (2022.01); G06V 10/82 (2022.01); G06V 20/52 (2022.01)
CPC G06F 3/017 (2013.01) [G06V 10/761 (2022.01); G06V 10/82 (2022.01); G06V 20/52 (2022.01)] 18 Claims
OG exemplary drawing
 
1. A computer-implemented method for gesture-based standard operating procedure (SOP) detection, the method comprising:
receiving a first set of images included in a first video captured by a camera that monitors a human performing a task;
processing the first set of images of the first video using a first machine learning (ML) model to determine whether the first set of images of the first video depicts a gesture that is included in a predefined set of gestures;
in response to determining that the first set of images of the first video depicts a gesture included in a predefined set of gestures, processing a second set of images included in the first video using a second ML model to determine a first gesture type of the gesture;
comparing the first gesture type with a first expected gesture type to determine whether performance of the task conforms to a standard operating procedure (SOP) for the task; and
providing feedback representative of a comparison result in a user interface,
wherein a number of images in the first set of images of the first video is less than a number of images in the second set of images of the first video.