CN112069898A - Method and device for recognizing human face group attribute based on transfer learning - Google Patents
Method and device for recognizing human face group attribute based on transfer learning Download PDFInfo
- Publication number
- CN112069898A CN112069898A CN202010775628.4A CN202010775628A CN112069898A CN 112069898 A CN112069898 A CN 112069898A CN 202010775628 A CN202010775628 A CN 202010775628A CN 112069898 A CN112069898 A CN 112069898A
- Authority
- CN
- China
- Prior art keywords
- face
- ethnic
- attribute
- image
- transfer learning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 62
- 238000013526 transfer learning Methods 0.000 title claims abstract description 51
- 238000012549 training Methods 0.000 claims description 41
- 238000007781 pre-processing Methods 0.000 claims description 26
- 238000012937 correction Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 abstract description 5
- 238000013145 classification model Methods 0.000 abstract description 4
- 238000000605 extraction Methods 0.000 description 10
- 238000001514 detection method Methods 0.000 description 7
- 238000012544 monitoring process Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007636 ensemble learning method Methods 0.000 description 1
- 230000008921 facial expression Effects 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 210000000697 sensory organ Anatomy 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention provides a method and a device for identifying human face group attributes based on transfer learning, wherein the identification method comprises the following steps: acquiring an image to be identified; and inputting the image into a pre-trained ethnic group attribute recognition model based on a transfer learning method, and calculating to obtain the human face ethnic group attribute in the image. According to the human face ethnic group attribute recognition method based on the transfer learning, the ethnic group attribute recognition model is trained by adopting the transfer learning-based method, so that the high-robustness ethnic group attribute recognition model can be obtained. When the human face ethnic group attribute is identified, the acquired image to be identified can be directly input into the ethnic group attribute classification model trained in advance for calculation, and the ethnic group attribute in the image to be identified is acquired, so that the identification method is efficient and reliable.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a method and a device for recognizing human face group attributes based on transfer learning.
Background
In recent years, with popularization of high-definition video monitoring systems in public places, effective technical assistance is provided for accident prevention and detection, missing population search and the like, but as the quantity of video monitoring data is multiplied, people in video monitoring are searched and analyzed in a manual mode, efficiency is low, and important information is easy to ignore. Therefore, the face attribute automatic identification under the monitoring scene is researched, the target pedestrian searching efficiency in monitoring can be obviously improved, and the method has great significance for reducing the monitoring cost and improving the safety level.
The face attribute recognition is to recognize the physiological characteristics (such as age, race, ethnic group, gender, complexion, and the like) of pedestrians and clothes (such as whether glasses, masks, hats and the like are worn) and can provide structured face attribute information through automatic recognition and analysis of the face attributes to effectively assist workers in searching personnel. In addition, as a multi-ethnic country, the face ethnic attribute recognition of China can effectively retain the face features of different ethnic nations and provide data for ethnic research in the future.
Disclosure of Invention
The invention provides a method and a device for identifying human face group attributes based on transfer learning, aiming at solving the technical problem of how to effectively identify the human face group attributes.
The method for identifying the human face ethnic group attribute based on the transfer learning comprises the following steps:
acquiring an image to be identified;
and inputting the image into a pre-trained ethnic group attribute recognition model based on a transfer learning method, and calculating to obtain the human face ethnic group attribute in the image.
According to the human face ethnic group attribute recognition method based on the transfer learning, the ethnic group attribute recognition model is trained by adopting the transfer learning-based method, and the high-robustness ethnic group attribute recognition model can be obtained. When the human face ethnic group attribute is identified, the acquired image to be identified can be directly input into the ethnic group attribute classification model trained in advance for calculation, and the ethnic group attribute in the image to be identified is acquired, so that the identification method is efficient and reliable.
According to some embodiments of the invention, a method of training the population attribute recognition model based on a transfer learning method comprises:
establishing a training network model, and adopting an ethnic face data set with ethnic attribute marks to perform pre-training to obtain an initial model;
and retraining the initial model by adopting a ethnic face data set with ethnic attribute marks to obtain the ethnic group attribute recognition model.
In some embodiments of the present invention, the data images in the ethnic and ethnic face datasets are preprocessed before model training using the ethnic and ethnic face datasets.
According to some embodiments of the invention, the identification method further comprises:
and after the image to be identified is obtained, preprocessing the image.
In some embodiments of the invention, the pre-processing comprises: image noise reduction, face recognition and face correction.
The device for identifying the human face ethnic group attribute based on the transfer learning comprises the following steps:
the acquisition unit is used for acquiring an image to be identified;
and the recognition unit is used for inputting the image into a pre-trained clan attribute recognition model based on a transfer learning method, and calculating to obtain the face clan attribute in the image.
According to the human face ethnic group attribute recognition device based on the transfer learning, the ethnic group attribute recognition model is trained by adopting a transfer learning-based method, and a high-robustness ethnic group attribute recognition model can be obtained. When the human face ethnic group attribute is identified, the image to be identified acquired by the acquisition module can be directly input into the ethnic group attribute model trained in advance for calculation, and the ethnic group attribute in the image to be identified is acquired.
According to some embodiments of the invention, the identification device further comprises: a model training unit for training the population attribute model, the model training unit being specifically configured to:
establishing a training network model, and adopting an ethnic face data set with ethnic attribute marks to perform pre-training to obtain an initial model;
and retraining the initial model by adopting a ethnic face data set with ethnic attribute marks to obtain the ethnic group attribute recognition model.
In some embodiments of the invention, the identifying means further comprises:
and the preprocessing unit is used for preprocessing the data images in the ethnic face data set and the ethnic face data set before model training is carried out by adopting the ethnic face data set and the ethnic face data set.
According to some embodiments of the invention, the pre-processing unit is further configured to: and after the acquisition unit acquires the image to be identified, preprocessing the image.
In some embodiments of the invention, the pre-processing comprises: image noise reduction, face recognition and face correction.
Drawings
Fig. 1 is a flowchart of a method for identifying a face population attribute based on transfer learning according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for identifying human face population attributes based on transfer learning according to an embodiment of the present invention;
FIG. 3 is a flowchart of a family property model training method according to an embodiment of the present invention;
FIG. 4 is a flowchart of a family property model training method according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of an apparatus for identifying human face population attributes based on transfer learning according to an embodiment of the present invention;
fig. 6 is a schematic diagram illustrating a recognition apparatus for face population attributes based on transfer learning according to an embodiment of the present invention.
Reference numerals:
the identification means 100 is arranged to identify the device,
the system comprises an acquisition unit 10, a recognition unit 20, a model training unit 30 and a preprocessing unit 40.
Detailed Description
To further explain the technical means and effects of the present invention adopted to achieve the intended purpose, the present invention will be described in detail with reference to the accompanying drawings and preferred embodiments.
Ethnicity attribute recognition generally refers to recognition of east asian race (yellow race), caucasian race (white race), nigromas (black race), australian race (brown race). In the related art, the ethnicity attribute identification mainly adopts the following method:
1. the three-dimensional face ethnicity is classified based on a three-dimensional point cloud solution. Although the three-dimensional face attribute recognition is slightly influenced by the face pose, the three-dimensional face data is usually difficult to acquire, and at present, the race recognition of the face is still performed by using the two-dimensional image.
2. PCA conversion is carried out on the center area of the five sense organs of the face, then special features are extracted by adopting an LDA analysis method, and the face ethnicity is classified by using an svm method.
3. Wavelet packet decomposition is carried out on the face image, decomposition coefficients are screened, a feature subspace of the face race is selected by an LDB method, and then the face race is classified by using svm.
4. And designing a generation network to generate a face race sample, and using the generated sample to improve the accuracy of face race classification.
Although the above method and model can also be applied to ethnic group identification, because people belonging to the same ethnic group can be divided into different ethnic groups, the discrimination of the human face features among ethnic groups is low compared with that of ethnic face features, and the method has poor effect on the identification of the human face ethnic group attributes.
Aiming at the identification of the attribute of the face clan, a face clan identification method based on region sparsity and weighted residual errors is adopted in the related technology; or normalizing the front face and the side face, extracting feature points by adopting a convolutional neural network, and calculating features on the basis of the feature points to identify the ethnic attribute of the face. The method comprises the steps of extracting artificial features of a human face, and then using a classifier to identify the attributes of the human face; or designing convolutional neural networks with different structures as classifiers and identifying the ethnic attributes of the face by adopting an ensemble learning method.
The method has strict requirements on the angle of the face and the quality of the input image, and when the quality of the input face image is poor, the recognition rate is low.
Aiming at the problems in the prior art, the invention provides a method and a device for identifying human face ethnic group attributes based on transfer learning.
As shown in fig. 1 and fig. 2, the method for identifying a face population attribute based on transfer learning according to an embodiment of the present invention includes:
s100, acquiring an image to be identified;
s200, inputting the image into a pre-trained clan attribute recognition model based on a transfer learning method, and calculating to obtain the human face clan attribute in the image.
According to the human face ethnic group attribute recognition method based on the transfer learning, the ethnic group attribute recognition model is trained by adopting the transfer learning-based method, and the high-robustness ethnic group attribute recognition model is obtained. When the human face ethnic group attribute is identified, the acquired image to be identified can be directly input into the ethnic group attribute classification model trained in advance for calculation, and the ethnic group attribute in the image to be identified is acquired, so that the identification method is efficient and reliable.
According to some embodiments of the present invention, as shown in fig. 3 and 4, the method for training the population property recognition model based on the transfer learning method includes:
a100, establishing a training network model, and pre-training by adopting an ethnic face data set with ethnic attribute marks to obtain an initial model;
it should be noted that the race attribute flag may be "caucasian", "yellow race", "black race", or "brown race", and the race face data set may adopt an existing public data set, so as to facilitate acquisition of the race face data set.
And A200, retraining the initial model by adopting a ethnic human face data set with ethnic attribute marks to obtain an ethnic attribute recognition model. The national attribute flag may be a national attribute such as "chinese", "hui nationality" or "white nationality".
In some embodiments of the present invention, the data images in the ethnic and ethnic face datasets are preprocessed before model training using the ethnic and ethnic face datasets. The preprocessing may include performing noise reduction processing, face recognition processing, face correction processing, and the like on the data image. By preprocessing the data image, the training speed of the model and the accuracy and reliability of the classification and identification of the model are improved.
Similarly, in the recognition process, after the image to be recognized is acquired, the image may be preprocessed in the same preprocessing manner.
As shown in fig. 5, an apparatus 100 for identifying a face population attribute based on transfer learning according to an embodiment of the present invention includes: an acquisition unit 10 and an identification unit 20.
The acquisition unit 10 is used to acquire an image to be recognized. The recognition unit 20 is configured to input the image into a population attribute recognition model trained in advance based on a transfer learning method, and calculate a face population attribute in the image.
According to the human face population attribute recognition device 100 based on the transfer learning, a population attribute recognition model is trained by adopting a transfer learning-based method, so that a high-robustness population attribute recognition model is obtained. When the human face ethnic group attribute is identified, the image to be identified acquired by the acquisition module can be directly input into the ethnic group attribute model trained in advance for calculation, and the ethnic group attribute in the image to be identified is acquired.
According to some embodiments of the invention, as shown in fig. 5, the identification apparatus 100 further comprises: the model training unit 30 is configured to train a population attribute model, and the model training unit 30 is specifically configured to:
establishing a training network model, and adopting an ethnic face data set with ethnic attribute marks to perform pre-training to obtain an initial model;
and retraining the initial model by adopting a ethnic face data set with ethnic attribute marks to obtain a ethnic group attribute recognition model.
In some embodiments of the present invention, as shown in fig. 5, the identification apparatus 100 further comprises: and the preprocessing unit 40 is configured to preprocess the data images in the ethnic face data set and the ethnic face data set before performing model training using the ethnic face data set and the ethnic face data set. The preprocessing may include performing noise reduction processing, face recognition processing, face correction processing, and the like on the data image. The data image is preprocessed through the preprocessing module, so that the training speed of the model and the accuracy and reliability of the classification and identification of the model are improved.
Similarly, the preprocessing unit 40 is also used for: after the acquisition unit 10 acquires the image to be recognized, the image is preprocessed in the same preprocessing manner.
The following describes a method and an apparatus for identifying face segment attributes based on transfer learning in detail with reference to the accompanying drawings in a specific embodiment. It is to be understood that the following description is only exemplary, and not a specific limitation of the invention.
In order to solve the problems of poor robustness and the like in the prior art, the invention provides a method and a device for identifying human face group attributes based on transfer learning. The invention adopts a transfer learning method, firstly uses an open data set to carry out and train a deep network model, constructs a Chinese ethnic human face data set, and uses transfer learning to classify human face ethnic groups on the data set, thereby realizing the attribute identification of the human face ethnic groups. The method has certain robustness on the quality of an input image, the expression and the posture of a human face, and improves the classification effect of human face ethnic group attributes by extracting depth features with ethnic group semantic information.
As shown in fig. 6, the identification device of the present invention includes: the system comprises an image data acquisition unit, a face detection and correction unit, a face group depth feature extraction unit and an attribute identification unit. Firstly, an image data acquisition unit is used for acquiring image data, then the acquired image data is subjected to face detection and correction, finally, the face is input into a trained face ethnic group depth feature extraction and attribute recognition unit, and a face ethnic group attribute recognition result is output.
Specifically, as shown in fig. 2, in the image data acquisition unit, the image data acquisition equipment is used for acquiring image data and preprocessing the image. In the face detection and correction unit, face detection and face direction correction are performed on the face contained in the image. And in the face group depth feature extraction and attribute identification unit, the corrected face is input to carry out depth feature extraction on the face, and the depth feature is utilized to carry out group attribute classification, so that the face group attribute identification is realized.
The preprocessing of the image comprises preprocessing of denoising and the like of the image of the human face. The mtcnn network model is specifically used for detecting the human face in the human face detection and correction unit, and the model can acquire the key point positions of the left eye and the right eye. During correction, the angles of the connecting line of the key points of the left eye and the right eye of the human face and the horizontal line are calculated, and the angle of the human face is rotated to enable the connecting line of the binocular of the human face to be parallel to the horizontal line. In the face group depth feature extraction and attribute unit, a face group classification depth network model is designed, the model comprises two parts of depth feature extraction and group attribute classification, the depth feature extraction can adopt a currently popular face recognition backbone network to carry out depth feature extraction, and in the group attribute classification, 2 full-connection layers and one softmax layer are used for classification.
The invention designs a method for classifying face ethnic groups based on transfer learning, which comprises the steps of firstly pre-training a face ethnic group classification network model by using a public face data set marked with ethnic attributes to obtain pre-training network model parameters, then constructing a Chinese ethnic face data set, finely adjusting the pre-trained network model by using the data set, and training to obtain a final face ethnic classification model. The flow chart is shown in fig. 4, and the specific training process is as follows:
and S1, correcting the face of the public face data set containing the race attributes, and inputting the public face data set to pre-train the face race network model, wherein in the process, the race attributes are used as face attribute labels to classify and recognize the race of the face, so that the pre-trained face race deep network model is obtained.
S2, constructing a Chinese ethnic face data set, wherein one training sample of the data set comprises a face photo and ethnic attribute labels thereof, and constructing the Chinese ethnic face data set by using a large number of different ethnic face training samples.
S3, the face group classification network model is initialized by using the pre-trained network weight in the step S1, and then the network parameters of the feature extraction part are fixed.
And S4, performing face detection and face correction on the Chinese ethnic group face data set constructed in the step S2, and training a face ethnic group classification network by using the corrected face photos until the model converges. The step can fine-tune 2 full-connection layers for classification to obtain a face ethnic group classification network model.
And in the face group depth feature extraction and attribute identification unit, inputting the face into the trained face group classification network to obtain a face group attribute identification result.
It should be noted that mtcnn, fully connected layer, softmax layer are prior art and will not be described in detail here.
In summary, the method and device for identifying the human face population attribute based on the transfer learning provided by the invention have the following advantages:
compared with the existing face ethnic group attribute recognition technology, the invention uses the public data set to pre-train the face ethnic group classification network, and then constructs the Chinese ethnic face database for transfer learning, so that the face ethnic group attribute classification precision can be effectively improved;
the invention uses the public data set to pre-train the ethnic group classification network, then uses the Chinese ethnic group face database to perform transfer learning, and has certain robustness on the quality of the input face image, the face expression and the posture.
While the invention has been described in connection with specific embodiments thereof, it is to be understood that it is intended by the appended drawings and description that the invention may be embodied in other specific forms without departing from the spirit or scope of the invention.
Claims (10)
1. A method for identifying human face ethnic group attribute based on transfer learning is characterized by comprising the following steps:
acquiring an image to be identified;
and inputting the image into a pre-trained ethnic group attribute recognition model based on a transfer learning method, and calculating to obtain the human face ethnic group attribute in the image.
2. The method for identifying human face population attributes based on transfer learning of claim 1, wherein the method for training the population attribute identification model based on the transfer learning method comprises:
establishing a training network model, and adopting an ethnic face data set with ethnic attribute marks to perform pre-training to obtain an initial model;
and retraining the initial model by adopting a ethnic face data set with ethnic attribute marks to obtain the ethnic group attribute recognition model.
3. The method of claim 2, wherein the data images in the ethnic face dataset and the ethnic face dataset are preprocessed before model training using the ethnic face dataset and the ethnic face dataset.
4. The method for identifying a human face population attribute based on transfer learning of claim 1, wherein the method further comprises:
and after the image to be identified is obtained, preprocessing the image.
5. The method for identifying a human face population attribute based on transfer learning according to claim 3 or 4, wherein the preprocessing comprises: image noise reduction, face recognition and face correction.
6. An apparatus for recognizing a face population attribute based on transfer learning, comprising:
the acquisition unit is used for acquiring an image to be identified;
and the recognition unit is used for inputting the image into a pre-trained clan attribute recognition model based on a transfer learning method, and calculating to obtain the face clan attribute in the image.
7. The apparatus for identifying a human face population attribute based on transfer learning of claim 6, wherein the apparatus further comprises: a model training unit for training the population attribute model, the model training unit being specifically configured to:
establishing a training network model, and adopting an ethnic face data set with ethnic attribute marks to perform pre-training to obtain an initial model;
and retraining the initial model by adopting a ethnic face data set with ethnic attribute marks to obtain the ethnic group attribute recognition model.
8. The apparatus for identifying a human face population attribute based on transfer learning of claim 7, wherein the apparatus further comprises:
and the preprocessing unit is used for preprocessing the data images in the ethnic face data set and the ethnic face data set before model training is carried out by adopting the ethnic face data set and the ethnic face data set.
9. The apparatus for identifying a face population attribute based on transfer learning of claim 8, wherein the preprocessing unit is further configured to: and after the acquisition unit acquires the image to be identified, preprocessing the image.
10. The apparatus for identifying a human face population attribute based on transfer learning of claim 9, wherein the preprocessing comprises: image noise reduction, face recognition and face correction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010775628.4A CN112069898A (en) | 2020-08-05 | 2020-08-05 | Method and device for recognizing human face group attribute based on transfer learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010775628.4A CN112069898A (en) | 2020-08-05 | 2020-08-05 | Method and device for recognizing human face group attribute based on transfer learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112069898A true CN112069898A (en) | 2020-12-11 |
Family
ID=73656447
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010775628.4A Pending CN112069898A (en) | 2020-08-05 | 2020-08-05 | Method and device for recognizing human face group attribute based on transfer learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112069898A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112612023A (en) * | 2020-12-14 | 2021-04-06 | 中国电子科技集团公司电子科学研究院 | Radar target identification method and computer readable storage medium |
| CN114360008A (en) * | 2021-12-23 | 2022-04-15 | 上海清鹤科技股份有限公司 | Generation method of face authentication model, authentication method, equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106295584A (en) * | 2016-08-16 | 2017-01-04 | 深圳云天励飞技术有限公司 | Depth migration study is in the recognition methods of crowd's attribute |
| CN109325443A (en) * | 2018-09-19 | 2019-02-12 | 南京航空航天大学 | A kind of face character recognition methods based on the study of more example multi-tag depth migrations |
-
2020
- 2020-08-05 CN CN202010775628.4A patent/CN112069898A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106295584A (en) * | 2016-08-16 | 2017-01-04 | 深圳云天励飞技术有限公司 | Depth migration study is in the recognition methods of crowd's attribute |
| CN107766787A (en) * | 2016-08-16 | 2018-03-06 | 深圳云天励飞技术有限公司 | Face character recognition methods, device, terminal and storage medium |
| CN109325443A (en) * | 2018-09-19 | 2019-02-12 | 南京航空航天大学 | A kind of face character recognition methods based on the study of more example multi-tag depth migrations |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112612023A (en) * | 2020-12-14 | 2021-04-06 | 中国电子科技集团公司电子科学研究院 | Radar target identification method and computer readable storage medium |
| CN114360008A (en) * | 2021-12-23 | 2022-04-15 | 上海清鹤科技股份有限公司 | Generation method of face authentication model, authentication method, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109409297B (en) | Identity recognition method based on dual-channel convolutional neural network | |
| US10445602B2 (en) | Apparatus and method for recognizing traffic signs | |
| CN111126240B (en) | Three-channel feature fusion face recognition method | |
| CN106570491A (en) | Robot intelligent interaction method and intelligent robot | |
| CN102902986A (en) | Automatic gender identification system and method | |
| WO2020164278A1 (en) | Image processing method and device, electronic equipment and readable storage medium | |
| CN104268586A (en) | Multi-visual-angle action recognition method | |
| CN111401303B (en) | Cross-visual angle gait recognition method with separated identity and visual angle characteristics | |
| CN106874877A (en) | A kind of combination is local and global characteristics without constraint face verification method | |
| Sharma et al. | Recognition of single handed sign language gestures using contour tracing descriptor | |
| CN110796101A (en) | Face recognition method and system of embedded platform | |
| CN109344909A (en) | A kind of personal identification method based on multichannel convolutive neural network | |
| CN113221655B (en) | Face spoofing detection method based on feature space constraint | |
| CN112069898A (en) | Method and device for recognizing human face group attribute based on transfer learning | |
| CN111666845A (en) | Small sample deep learning multi-mode sign language recognition method based on key frame sampling | |
| CN112836675A (en) | Unsupervised pedestrian re-identification method and system based on clustering-generated pseudo label | |
| US9135562B2 (en) | Method for gender verification of individuals based on multimodal data analysis utilizing an individual's expression prompted by a greeting | |
| CN113158828B (en) | Facial emotion calibration method and system based on deep learning | |
| CN108596057B (en) | Information security management system based on face recognition | |
| CN110110606A (en) | The fusion method of visible light neural network based and infrared face image | |
| CN111582195B (en) | Construction method of Chinese lip language monosyllabic recognition classifier | |
| CN113159002A (en) | Facial expression recognition method based on self-attention weight auxiliary module | |
| CN111950452A (en) | Face recognition method | |
| CN112613471A (en) | Face living body detection method and device and computer readable storage medium | |
| Hsiao et al. | Iris location and recognition by deep-learning networks based design for biometric authorization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201211 |
|
| RJ01 | Rejection of invention patent application after publication |