Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
EMOTION RECOGNITION METHOD AND APPARATUS, DEVICE, AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2024/000867
Kind Code:
A1
Abstract:
An emotion recognition method and apparatus, a device, and a storage medium, applied in the technical field of neural networks. A method for training an emotion recognition model comprises: acquiring video to be tested and audio to be tested (S201); determining multiple video frames to be tested in the video to be tested, and generating text data to be tested corresponding to emotion labels by splicing the emotion labels in a label set into a template for text to be tested (S202); inputting the video frames to be tested, the text data to be tested, and the audio to be tested into the emotion recognition model, to obtain text encoded data corresponding to the non-text encoded data to be tested, and the text encoded data to be tested (S203); generating similarity data to be tested by using the non-text encoded data to be tested and the respective text encoded data to be tested (S204); determining an emotion label corresponding to maximum similarity data to be tested to be an emotion recognition result corresponding to the video to be tested (S205). The present method introduces semantic information contained in a label, so that the accuracy is increased.

Inventors:
ZHANG RUNZE (CN)
LI RENGANG (CN)
ZHAO YAQIAN (CN)
GUO ZHENHUA (CN)
FAN BAOYU (CN)
LI XIAOCHUAN (CN)
Application Number:
PCT/CN2022/121852
Publication Date:
January 04, 2024
Filing Date:
September 27, 2022
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
INSPUR ELECTRONIC INFORMATION INDUSTRY CO LTD (CN)
International Classes:
G06V40/16; G06V10/74; G06V10/82
Foreign References:
CN115050077A2022-09-13
CN113920561A2022-01-11
CN114550057A2022-05-27
CN113536999A2021-10-22
CN112926525A2021-06-08
CN111339913A2020-06-26
US20200286506A12020-09-10
Attorney, Agent or Firm:
BEIJING RUN ZEHENG INTELLECTUAL PROPERTY LAW FIRM (CN)
Download PDF: