US 12,406,532 B2
Behavior recognition artificial intelligence network system and method for efficient recognition of hand signals and gestures
Kyoung Won Min, Seongnam-si (KR); Ganzorig Gankhuyag, Seoul (KR); Haeng Seon Son, Seongnam-si (KR); Seon Young Lee, Seoul (KR); Young Bo Shim, Seongnam-si (KR); and Chang Gue Park, Seoul (KR)
Assigned to Korea Electronics Technology Institute, Seongnam-si (KR)
Filed by Korea Electronics Technology Institute, Seongnam-si (KR)
Filed on Dec. 1, 2022, as Appl. No. 18/073,058.
Claims priority of application No. 10-2021-0183642 (KR), filed on Dec. 21, 2021.
Prior Publication US 2023/0196841 A1, Jun. 22, 2023
Int. Cl. G06V 40/20 (2022.01); G06V 10/22 (2022.01); G06V 10/40 (2022.01)
CPC G06V 40/28 (2022.01) [G06V 10/22 (2022.01); G06V 10/40 (2022.01)] 12 Claims
OG exemplary drawing
 
1. A behavior recognition method for recognition of hand signals and gestures, the method comprising:
extracting, by a behavior recognition AI network system, key points F1 from bounding box data of an object which makes hand signals to be inputted by sequence, and generating skeleton data of the object;
calculating, by the behavior recognition AI network system, a respective length F2 and a respective angle F3 of a respective bone vector based on the key points F1 and the skeleton data, and extracting spatial features from a result of calculating, each spatial feature including the respective length F2 and the respective angle F3 of the respective bone vector;
generating a new feature F4 by concatenating the input feature of the key points F1 and the spatial features each including the respective length F2 and the respective angle F3 of the respective bone vector; and
recognizing hand signals made by the object or a behavior of the object by inputting the new feature F4 to a neural network-based model.