US 12,469,295 B2
Method and device for acquiring information, and system for detecting attention degree
Mingjun Nie, Beijing (CN); Jingjing Wang, Beijing (CN); Jiangxu Wu, Beijing (CN); Miaofeng Hu, Beijing (CN); Shiwei Ma, Beijing (CN); and Ran Zhang, Beijing (CN)
Assigned to BEIJING WODONG TIANJUN INFORMATION TECHNOLOGY CO., LTD., Beijing (CL); and BEIJING JINGDONG CENTURY TRADING CO., LTD, Beijing (CN)
Appl. No. 17/999,971
Filed by BEIJING WODONG TIANJUN INFORMATION TECHNOLOGY CO., LTD., Beijing (CN); and BEIJING JINGDONG CENTURY TRADING CO., LTD., Beijing (CN)
PCT Filed May 13, 2021, PCT No. PCT/CN2021/093517
§ 371(c)(1), (2) Date Nov. 28, 2022,
PCT Pub. No. WO2021/238664, PCT Pub. Date Dec. 2, 2021.
Claims priority of application No. 202010479278.7 (CN), filed on May 29, 2020.
Prior Publication US 2023/0206642 A1, Jun. 29, 2023
Int. Cl. G06V 20/52 (2022.01); G06F 16/215 (2019.01); G06F 16/28 (2019.01); G06V 10/70 (2022.01); G06V 20/40 (2022.01); G06V 40/10 (2022.01); G06V 40/20 (2022.01)
CPC G06V 20/52 (2022.01) [G06F 16/215 (2019.01); G06F 16/285 (2019.01); G06V 10/70 (2022.01); G06V 20/46 (2022.01); G06V 40/10 (2022.01); G06V 40/20 (2022.01)] 9 Claims
OG exemplary drawing
 
1. An attention degree detecting system, comprising:
a video capturing equipment configured to acquire video information of a predetermined area;
an information acquiring apparatus comprising:
a first memory configured to store instructions;
a first processor coupled to the first memory, wherein based on the instructions stored in the first memory, the first processor is configured to:
acquire a video frame captured by the video capturing equipment;
detect whether a pedestrian exists in the video frame by using a first deep-learning model;
process the video frame by using a second deep-learning model to generate a pedestrian bounding box of the pedestrian under a condition that the pedestrian exists in the video frame, and assign a corresponding pedestrian identifier to the pedestrian bounding box;
identify an image in the pedestrian bounding box by using a third deep-learning model to identify feature information of the pedestrian, wherein the feature information of the pedestrian comprises behavior feature information of the pedestrian, and the behavior feature information of the pedestrian comprises at least one of taking, putting back or holding; and
write a pedestrian historical data into a database, wherein the pedestrian historical data comprises a timestamp of the video frame, an identifier of the video frame, coordinate information of the pedestrian bounding box, the pedestrian identifier and the feature information of the pedestrian; and
an attention degree detecting apparatus comprising:
a second memory configured to store instructions;
a second processor coupled to the second memory, wherein based on the instructions stored in the second memory, the second processor is configured to:
receive a query request for querying an attention degree of a predetermined area;
extract a plurality of pedestrian historical data associated with the predetermined area from a predetermined database;
extract a plurality of pedestrian historical data to be detected from the plurality of pedestrian historical data, wherein timestamps in the plurality of pedestrian historical data to be detected are within a predetermined time range;
count the number of pieces of feature information of the pedestrians in the pedestrian historical data to be detected; and
determine the attention degree of the predetermined area within the predetermined time range according to the result of counting.