IP Library Granted Patent US 12,236,704
Granted Patent B2
US 12,236,704 · App. 17/618,974 · Granted Feb 25, 2025

Person state detection apparatus, person state detection method, and non-transitory computer readable medium storing program

Inventors: Noboru Yoshida (Tokyo, JP); Shoji Nishimura (Tokyo, JP)
Assignee: NEC CORPORATION
G06V40/103G06T7/73
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,236,704
App. No.
17/618,974
Granted
Feb 25, 2025
Kind
B2
Abstract

A person state detection apparatus ( 10 ) according to the present disclosure includes an acquisition unit ( 11 ) for acquiring a two-dimensional image obtained by capturing a person, a skeletal structure detection unit ( 12 ) for detecting a two-dimensional skeletal structure of the person based on the acquired two-dimensional image, an estimation unit ( 13 ) for estimating a height of the person standing upright in a two-dimensional image space based on the detected two-dimensional skeletal structure, and a state detection unit ( 14 ) for detecting the state of the person based on a height of an area where the person is present in the two-dimensional image.

Claims (51)

1. A person state detection apparatus comprising:

at least one memory storing instructions, and

at least one processor configured to execute the instructions stored in the at least one memory to;

acquire an image obtained by capturing a person;

detect a skeletal structure of the person based on the acquired image;

estimate a height of the person standing upright in a two-dimensional image space based on the detected skeletal structure; and

detect a state, which includes at least one of a posture or an action, of the person based on the estimated height of the person standing upright and a height of an area where the person is present in the image.

2. The person state detection apparatus according to claim 1 , wherein

the at least one processor is further configured to execute the instructions stored in the at least one memory to detect the state of the person based on a ratio of the height of the person standing upright to the height of the area where the person is present.

3. The person state detection apparatus according to claim 2 , wherein

the at least one processor is further configured to execute the instructions stored in the at least one memory to detect that the person is crouching down based on a result of a comparison between the ratio and a predetermined threshold.

4. The person state detection apparatus according to claim 2 , wherein

the at least one processor is further configured to execute the instructions stored in the at least one memory to detect that the person is lying down based on a result of a comparison between the ratio and a predetermined threshold.

5. The person state detection apparatus according to claim 1 , wherein

the at least one processor is further configured to execute the instructions stored in the at least one memory to estimate the height of the person standing upright based on a length of a bone in an image space included in the skeletal structure.

6. The person state detection apparatus according to claim 5 , wherein

the at least one processor is further configured to execute the instructions stored in the at least one memory to estimate the height of the person standing upright based on a sum of the lengths of the bones from a foot to a head included in the skeletal structure.

7. The person state detection apparatus according to claim 5 , wherein

the at least one processor is further configured to execute the instructions stored in the at least one memory to estimate the height of the person standing upright based on a skeleton model showing a relationship between the length of the bone and a length of a whole body of the person in the image space.

8. The person state detection apparatus according to claim 1 , wherein

the at least one processor is further configured to execute the instructions stored in the at least one memory to estimate the height of the person standing upright based on a three-dimensional skeleton model fitted to the skeletal structure based on the imaging parameter of the image.

9. The person state detection apparatus according to claim 1 , wherein

the at least one processor is further configured to execute the instructions stored in the at least one memory to detect the state of the person based on a result of a comparison between a predetermined threshold and a ratio of the height of the person standing upright to the height of the area where the person is present.

10. A person state detection method comprising:

acquiring an image obtained by capturing a person;

detecting a skeletal structure of the person based on the acquired image;

estimating a height of the person standing upright in an image space based on the detected skeletal structure; and

detecting a state, which includes at least one of a posture or an action, of the person based on the estimated height of the person standing upright and a height of an area where the person is present in the image.

11. The person state detection method according to claim 10 , wherein

the person state detection method further comprises detecting the state of the person based on a ratio of the height of the person standing upright to the height of the area where the person is present.

12. The person state detection method according to claim 11 , wherein

the person state detection method further comprises detecting that the person is crouching down based on a result of a comparison between the ratio and a predetermined threshold.

13. The person state detection method according to claim 11 , wherein

the person state detection method further comprises detecting that the person is lying down based on a result of a comparison between the ratio and a predetermined threshold.

14. The person state detection method according to claim 10 , wherein

the person state detection method further comprises estimating the height of the person standing upright based on a length of a bone in an image space included in the skeletal structure.

15. The person state detection method according to claim 14 , wherein

the person state detection method further comprises estimating the height of the person standing upright based on a sum of the lengths of the bones from a foot to a head included in the skeletal structure.

16. The person state detection method according to claim 14 , wherein

the person state detection method further comprises estimating the height of the person standing upright based on a skeleton model showing a relationship between the length of the bone and a length of a whole body of the person in the image space.

17. The person state detection method according to claim 10 , wherein

the person state detection method further comprises estimating the height of the person standing upright based on a three-dimensional skeleton model fitted to the skeletal structure based on the imaging parameter of the image.

18. The person state detection method according to claim 10 , wherein

the person state detection method further comprises detecting the state of the person based on a result of a comparison between a predetermined threshold and a ratio of the height of the person standing upright to the height of the area where the person is present.

19. A non-transitory computer readable medium storing a program for causing a computer to execute processing of:

acquiring an image obtained by capturing a person;

detecting a skeletal structure of the person based on the acquired image;

estimating a height of the person standing upright in an image space based on the detected skeletal structure; and

detecting a state, which includes at least one of a posture or an action, of the person based on the estimated height of the person standing upright and a height of an area where the person is present in the image.

20. The non-transitory computer readable medium according to claim 19 , wherein

the program further causes the computer to execute processing of detecting the state of the person based on a result of a comparison between a predetermined threshold and a ratio of the height of the person standing upright to the height of the area where the person is present.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 14, 2021
From: YOSHIDA, NOBORU; NISHIMURA, SHOJI
To: NEC CORPORATION
Reel/Frame 058381/0601 →
Continuity (1)
Related Publication 20220366716A1 · Nov 17, 2022
References Cited (17)
US 9739599B2 · Srinivasan · 2017 [cited by examiner]
US 10002460B2 · Black · 2018 [cited by examiner]
US 10482613B2 · Radwin · 2019 [cited by examiner]
US 11837006B2 · Dong · 2023 [cited by examiner]
US 20120281918A1 · Song · 2012 [cited by examiner]
US 20150213653A1 · Kord · 2015 [cited by examiner]
US 20200285846A1 · Oami · 2020 [cited by applicant]
US 20240193809A1 · Ostadabbas · 2024 [cited by examiner]
JP 2008146583A · 2008 [cited by applicant]
JP 2010237873A · 2010 [cited by applicant]
JP 2012120647A · 2012 [cited by applicant]
JP 2017503225A · 2017 [cited by applicant]
JP 6529062B · 2019 [cited by applicant]
WO 2019064375A1 · 2019 [cited by applicant]
International Search Report for PCT Application No. PCT/JP2019/025276, mailed on Sep. 17, 2019. [cited by applicant]
Zhe Cao et al., “Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields”, The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 7291-7299. [cited by applicant]
Indian Office Action for IN Application No. 202117060716 mailed on Jul. 4, 2022. [cited by applicant]