Unlock instant, AI-driven research and patent intelligence for your innovation.

Face recognition method and device based on video

A video and face technology, applied in the field of video-based face identification methods and devices, can solve problems such as low efficiency of face identification

Active Publication Date: 2022-04-22
TENCENT TECH (SHENZHEN) CO LTD +1
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] For this reason, the present invention proposes a method and device for face identification based on video, which uses a semi-supervised learning method to perform face identification and solves the technical problem of low efficiency of face identification in the prior art

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Face recognition method and device based on video
  • Face recognition method and device based on video
  • Face recognition method and device based on video

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0073] Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary and are intended to explain the present invention and should not be construed as limiting the present invention.

[0074] The video-based face identification method and device of the embodiments of the present invention are described below with reference to the accompanying drawings.

[0075] figure 1 It is a schematic flowchart of a video-based face identification method provided by an embodiment of the present invention.

[0076] Step 101, face recognition is performed on the video to obtain a face sequence.

[0077] Wherein, the face sequence is used to indicate the corresponding relationship between the recognized face image and the tim...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The present invention proposes a video-based face identification method and device, wherein the method includes: obtaining a face sequence by performing face recognition on a video, and obtaining a person and the moment when the person appears in the video After the corresponding relationship, use the corresponding relationship between the person and the time when the person appears in the video to mark the part of the sequence in the face sequence. The marked face sequence and the unmarked face sequence are used as training samples to conduct semi-supervised learning on the learning model. According to the learning results, the face images are identified by human annotation in the video. Due to the automatic labeling method, human labeling is saved, and at the same time, the labeled face sequence and the unlabeled face sequence are used as training samples for semi-supervised learning, which improves the accuracy of the labeling and solves the problem of face recognition in the prior art. Identify technical issues that are less efficient.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a video-based face recognition method and device. Background technique [0002] In the prior art, it is already possible to use movie scripts to identify characters appearing in movie videos, specifically, to use names of people to tag appearing face images. In the prior art, in order to realize this kind of face identification, there are two ways to realize it: [0003] One way of implementation is an unsupervised method, which needs to cluster the faces appearing in the video to obtain multiple clusters, and then analyze the time distribution of each cluster in the entire video, and compare it with the appearance of names in the movie script. Align the time distribution of the face recognition results. However, the accuracy of identification in this way depends on the result of clustering, resulting in an unsupervised method with low accuracy. [0004] Anoth...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06V40/16G06V10/774G06V10/764G06V10/766G06K9/62
CPCG06V40/172G06F18/24G06F18/214
Inventor 张一帆
Owner TENCENT TECH (SHENZHEN) CO LTD