US 11,954,942 B2
Human behavior recognition system and method using hierarchical class learning considering safety
Junghyun Cho, Seoul (KR); Ig Jae Kim, Seoul (KR); and Hochul Hwang, Seoul (KR)
Assigned to KOREA INSTITUTE OF SCIENCE AND TECHNOLOGY, Seoul (KR)
Filed by KOREA INSTITUTE OF SCIENCE AND TECHNOLOGY, Seoul (KR)
Filed on Dec. 30, 2021, as Appl. No. 17/565,453.
Claims priority of application No. 10-2020-0189449 (KR), filed on Dec. 31, 2020.
Prior Publication US 2022/0207920 A1, Jun. 30, 2022
Int. Cl. G06V 40/20 (2022.01); G06V 10/82 (2022.01); G06V 10/84 (2022.01)
CPC G06V 40/23 (2022.01) [G06V 10/82 (2022.01); G06V 10/84 (2022.01)] 21 Claims
OG exemplary drawing
 
1. A human behavior recognition system using hierarchical class learning considering safety, the human behavior recognition system comprising:
a behavior class definer configured to form a plurality of behavior classes by sub-setting a plurality of images each comprising a subject according to pre-designated behaviors and assign a behavior label to the plurality of images, wherein each behavior class comprises images with same or similar behaviors of the subject, and a same behavior label is assigned to images included in a same behavior class;
a safety class definer configured to calculate a safety index for the plurality of images, form a plurality of safety classes by sub-setting the plurality of images based on the safety index, and additionally assign a safety label to the plurality of images, wherein an object different from the subject in the image is recognized and the safety index is evaluated based on a pose of the recognized object with respect to the subject; and
a trainer configured to train a human recognition model by using the plurality of images defined as hierarchical classes by assigning the behavior label and the safety label as training images.