The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.

The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.

Date of Patent:
Jan. 13, 2026

Filed:

Apr. 18, 2024
Applicant:

Chung Ang University Industry Academic Cooperation Foundation, Seoul, KR;

Inventors:

Joon Ki Paik, Seoul, KR;

Ha Sil Park, Seoul, KR;

Joong Chol Shin, Seoul, KR;

Jin Sol Ha, Seoul, KR;

Attorney:
Primary Examiner:
Int. Cl.
CPC ...
G06V 40/20 (2022.01); G06V 10/764 (2022.01); G06V 10/77 (2022.01); G06V 10/80 (2022.01); G06V 10/82 (2022.01);
U.S. Cl.
CPC ...
G06V 40/20 (2022.01); G06V 10/764 (2022.01); G06V 10/7715 (2022.01); G06V 10/806 (2022.01); G06V 10/82 (2022.01);
Abstract

A method using deep learning includes: sampling video to generate a video clip composed sampled frames, and generating a set of differential images between the sampled frames; extracting a first feature map with spatial features and a second feature map with temporal features by applying the sampled frames and the set of differential images to a learned deep learning-based behavior recognition model, and generating a spatio-temporal combined feature map for each sampled frame by combining the first feature map and the second feature map; calculating a bi-directional exponential moving average using the spatio-temporal combined feature map, and then adjusting weights of the spatio-temporal combined feature maps; generating a temporal relationship importance map by applying the video clip and performing late fusion by applying the weight-adjusted spatio-temporal combined feature map and the temporal relationship importance map to the learned deep learning-based behavior recognition model, and classifying the behavior.


Find Patent Forward Citations

Loading…