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

Face detection method and device and electronic equipment

A face detection and to-be-detected technology, applied in the field of image processing, can solve the problems of large number of intermediate result channels, decreased data processing efficiency, difficulty in improving detection accuracy, etc. Effect

Pending Publication Date: 2021-12-31
NANJING HUAJIE IMI TECH CO LTD +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] There are many methods of face detection in the prior art. In order to improve the accuracy of face detection, it has been proposed to combine color images and depth images for face detection. However, convolutional neural networks are usually used for color images and depth images. The intermediate results obtained when the depth image is processed are fused, but the number of channels of the intermediate results is large, so there are many parameters that need to be fused, which will lead to a decrease in data processing efficiency, and the features of the color image and the channels of the features of the depth image The meanings are different, and it is difficult to achieve the purpose of improving the detection accuracy by directly integrating the two

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 detection method and device and electronic equipment
  • Face detection method and device and electronic equipment
  • Face detection method and device and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach 1

[0103] Embodiment 1: By using a convolutional neural network and using different receptive fields, features of different scales are extracted, for example, features of 5 scales can be extracted.

Embodiment approach 2

[0104] Embodiment 2: Using a convolutional neural network and using different receptive fields, multiple features of different scales are extracted, and the extracted features are down-sampled at least once to obtain features of other scales.

[0105] For example, through the convolutional neural network, in the case of using different receptive fields, the extraction includes the first feature, the second feature, and the third feature, wherein the scales of the first feature, the second feature, and the third feature are different, and the The third feature is down-sampled twice to obtain the fourth feature and the fifth feature.

[0106] There may be various manners of fusing features of multiple scales, which are not limited in this embodiment. Preferably, a BiFPN algorithm may be used.

[0107] In this example, refer to image 3 , shows the processing process of the face detection model, the color image and the depth image are input into the face detection model, the mod...

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 invention discloses a face detection method and device and electronic equipment, and the method comprises the steps: extracting the features of a to-be-detected color image and a to-be-detected depth image through a face detection model, and carrying out the analysis of the features of the color image and the features of the depth image; obtaining a classification vector and a position vector of a color image and a classification vector and a position vector of a depth image, fusing the classification vectors of the color image and the depth image, fusing the position vectors of the color image and the depth image, and outputting a final classification vector and a final position vector; and determining candidate face frame through the relationship between the classification vector and a threshold value, and then determining a target face frame representing a detection result based on the candidate face frame. Therefore, face detection is carried out by fully utilizing the characteristics of the color image and the depth image, and the respective detection results are fused, so that the data processing efficiency is improved, and the accuracy of face detection is improved.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a face detection method, device and electronic equipment. Background technique [0002] Face detection is a very important step in the face recognition process. Its purpose is to find the position coordinates of the minimum bounding box of the face in the input image or image group, and pass the position coordinates to the next process for subsequent detection and recognition. Work to lay the groundwork. [0003] There are many methods of face detection in the prior art. In order to improve the accuracy of face detection, it has been proposed to combine color images and depth images for face detection. However, convolutional neural networks are usually used for color images and depth images. The intermediate results obtained when the depth image is processed are fused, but the number of channels of the intermediate results is large, so there are many parameters that need to be fu...

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 Applications(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06F18/214G06F18/253
Inventor 封洁轩李骊
Owner NANJING HUAJIE IMI TECH CO LTD