Face recognition method based on cosine loss function

A loss function, face recognition technology, applied in the field of face recognition, can solve problems such as low training efficiency and inability to apply face recognition training, and achieve the effect of improving precision and accuracy and improving efficiency

Pending Publication Date: 2021-12-17
四川天翼网络股份有限公司
View PDF9 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although this technical solution overcomes the differences between training and testing by combining SoftMax loss, cosine similarity loss, and Center Loss as the objective function, this method is not suitable for face recognition training with a large amount of sample data. not tall
[0004] For example, the patent application with the application number CN202010188585.X discloses a face recognition method based on cosine loss with unconstrained conditions, which includes the following steps, S1, acquiring an image to be recognized, performing multi-scale transformation on the image to be recognized, and obtaining an image pyramid; S2 1. The image pyramid obtained in step S1 is input to the MTCNN network, and after the MTCNN network processes the image, facial feature points are obtained; S3, according to the facial feature points of step S2, face correction is performed; S4, using the processed data of step S3 Training the Inception-ResnetV1 convolutional neural network, although this technical solution uses the cosine loss function as a supervisory signal to train the classifier model, this method cannot be applied to face recognition training with a large amount of sample data, and the training efficiency is not high

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 based on cosine loss function
  • Face recognition method based on cosine loss function
  • Face recognition method based on cosine loss function

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 2

[0041] In this embodiment, a further improvement is made on the basis of the method provided in Embodiment 1. The cosine loss function designed in Embodiment 1 can make the angle between the face vector codes corresponding to the same person as small as possible, and make the face vector codes of different faces angle as large as possible.

[0042] Therefore, in this implementation, during the training process, the Embedding output by the CNN automatically matches the eigenvector encoding solved by the Thomson Problem.

[0043] Specifically, in step 4, using the cosine loss function to perform the nearest matching calculation process on the first face feature code and the second face feature code includes: respectively inputting the first face feature code and the second face feature code into the designed cosine In the loss function, the included angle θ1 of the first face feature code and the included angle θ2 of the second face feature code are calculated respectively; acco...

Embodiment 3

[0049] In the present embodiment, a kind of face recognition method based on cosine loss function comprises the following steps:

[0050] Step 1: Preprocessing the face image data that needs face recognition training to form a training sample set;

[0051] Step 2: Construct a CNN network, input the training sample set into the CNN network for face recognition convolutional neural network training, and output the first face feature code after the training is completed;

[0052] Step 3: Construct the Thomson Problem model, input the training sample set into the Thomson Problem model for solving, and calculate the second facial feature code;

[0053] Step 4: Design the cosine loss function, and use the cosine loss function to perform the nearest matching calculation on the first face feature encoding and the second face feature encoding, and the calculated nearest matching result is used to correct the parameters of the CNN network model until the calculated When the latest matc...

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 recognition method based on a cosine loss function. The face recognition method comprises the following steps: preprocessing face image data needing face recognition training to form a training sample set; constructing a CNN network, inputting the training sample set into the CNN network to carry out face recognition convolutional neural network training, and outputting a first face feature code after the training is completed; inputting the training sample set into a Thomson Problem model for solving, and calculating a second face feature code; performing recent matching calculation on the first face feature code and the second face feature code, correcting CNN model parameters according to a recent matching result, and outputting a face recognition result until the recent matching result meets a set threshold value. In the face recognition process, the coding vector of the face feature is adopted to replace the original FC, so that the training of the neural network is not influenced by the size of the training data set, and the face recognition efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to a face recognition method based on a cosine loss function. Background technique [0002] With the blessing of big data, more and more data are available for face recognition training, ranging from tens of thousands to hundreds of thousands to millions. The increase in data has brought us new problems. Large data sets are difficult to train on the current mainstream Loss functions (such as the Softmax series), because this type of loss function requires the last fully connected layer during training. (FC) The number of neurons should correspond to the number of people in the data set. When the number of people trained is hundreds of thousands, the parameter quantity of the last layer is often about 10 times higher than the previous parameter quantity, and the last layer may need to Occupying nearly 1G of video memory, the input BatchSize of such a 12G video card can only...

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): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/22G06F18/214
Inventor 刘栓
Owner 四川天翼网络股份有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products