US 12,265,930 B2
Work estimation apparatus, work estimation method, and non-transitory computer-readable storage medium
Keisuke Nishimura, Fujisawa (JP); Atsushi Wada, Shiraoka (JP); Masatoshi Takano, Yokohama (JP); Yuta Shirakawa, Kawasaki (JP); Yojiro Tonouchi, Inagi (JP); and Shojun Nakayama, Yokohama (JP)
Assigned to KABUSHIKI KAISHA TOSHIBA, Tokyo (JP)
Filed by KABUSHIKI KAISHA TOSHIBA, Tokyo (JP)
Filed on Mar. 8, 2022, as Appl. No. 17/653,931.
Claims priority of application No. 2021-154586 (JP), filed on Sep. 22, 2021.
Prior Publication US 2023/0103948 A1, Apr. 6, 2023
Int. Cl. G06T 7/70 (2017.01); G06Q 10/06 (2023.01); G06Q 10/0631 (2023.01); G06Q 10/0639 (2023.01); G06Q 10/067 (2023.01); G06T 7/73 (2017.01); G06T 11/20 (2006.01)
CPC G06Q 10/063114 (2013.01) [G06Q 10/06398 (2013.01); G06T 7/73 (2017.01); G06T 11/206 (2013.01); G06T 2207/10016 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30196 (2013.01)] 29 Claims
OG exemplary drawing
 
15. A work estimation method comprising:
acquiring video data on a predetermined area;
estimating a work state of a worker included in the video data based on the video data;
outputting display data including a graph related to the work state;
estimating the work state based on transition of positions and states of a plurality of body parts of the worker in a plurality of consecutive frames of the video data;
estimating the positions and the states of the plurality of body parts by applying a skeleton estimation model to each of the plurality of frames; and
estimating the work state at each time of the video data based on time-series data related to the positions and the states of the plurality of body parts,
wherein the positions and the states of the plurality of body parts include information on at least one of an eye position, a shoulder position, an elbow position, a wrist position, a waist position, a neck bending state, and a waist bending state,
wherein the positions and the states of the plurality of body parts include information on the wrist position, and the method further comprises:
performing an area determination of whether or not the wrist position is included in a particular range in the predetermined area, and
estimating the work state by using at least a determination result of the area determination.