Title:
IMAGE DETECTION METHOD, MODEL TRAINING METHOD, APPARATUS, DEVICE, AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2021/179852
Kind Code:
A1
Abstract:
The present application provides an image detection method, a model training method, an apparatus, a device, and a storage medium. The image detection method comprises: acquiring n face images; performing normalization processing on the n face images to obtain n input images; processing the mth input image by means of an attention module to obtain an attention image; processing the mth input image by means of a multi-scale encoder to obtain first feature information; processing the first feature information by means of a feature pyramid module to obtain second feature information; processing the attention image and the second feature information by means of a multi-scale decoding predictor to obtain indication information of a face key point corresponding to the mth input image; determining, according to indication information of face key points corresponding to the n input images, indication information of eye key points corresponding to the n input images; and determining a target drowsiness state according to the indication information of the eye key points respectively corresponding to the n input images.
Inventors:
LIU YUAN (CN)
Application Number:
PCT/CN2021/074779
Publication Date:
September 16, 2021
Filing Date:
February 02, 2021
Export Citation:
Assignee:
GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP LTD (CN)
International Classes:
G06K9/00; G06K9/62; G06N3/04; G06N3/08
Foreign References:
| CN111382714A | 2020-07-07 | |||
| CN110222607A | 2019-09-10 | |||
| CN109614876A | 2019-04-12 | |||
| CN110796147A | 2020-02-14 | |||
| KR20120113317A | 2012-10-15 |
Attorney, Agent or Firm:
BEIJING SAN GAO YONG XIN INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF: