CN111967381B - Face image quality grading and labeling method and device - Google Patents
Face image quality grading and labeling method and device Download PDFInfo
- Publication number
- CN111967381B CN111967381B CN202010822252.8A CN202010822252A CN111967381B CN 111967381 B CN111967381 B CN 111967381B CN 202010822252 A CN202010822252 A CN 202010822252A CN 111967381 B CN111967381 B CN 111967381B
- Authority
- CN
- China
- Prior art keywords
- face
- ith
- face image
- quality
- image
- 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.)
- Active
Links
- 238000002372 labelling Methods 0.000 title claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 37
- 239000013598 vector Substances 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims abstract description 19
- 238000013441 quality evaluation Methods 0.000 claims abstract description 18
- 230000009466 transformation Effects 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 25
- 238000012545 processing Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000013077 scoring method Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims 1
- 238000011156 evaluation Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000001303 quality assessment method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 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 marking face image quality scores, wherein the method comprises the following steps: the face image to be detected in the set and the external noise face image are processed through a face recognition network to obtain corresponding face characteristic vectors; wherein, each person at least comprises a reference picture; calculating an indicative function according to the reference graph characteristics of the ith person and the reference graph characteristics of other different persons; calculating a quality value for the kth map of the ith individual based on the indicative function; performing linear transformation on the quality value to obtain a final label value of the kth image to be tested of the ith person; and performing face quality evaluation model training by using the final label value. The method comprehensively considers the intra-class distance and the inter-class distance of the human face features and carries out the task of labeling the human face image training sample for the quality evaluation task. By applying the method, more accurate face image quality label values can be obtained, and a foundation is laid for better face quality evaluation networks for subsequent training.
Description
Technical Field
One or more embodiments of the invention relate to the technical field of image processing, in particular to a method and a device for marking face image quality scores.
Background
This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein may include concepts that could be pursued, but are not necessarily ones that have been previously conceived or pursued. Therefore, unless otherwise indicated herein, what is described in this section is not prior art to the description and claims in this application and is not admitted to be prior art by inclusion in this section.
With the rapid development of society and science and technology, face recognition technology is more and more widely applied and popularized, and for example, the face recognition technology is gradually applied to important security ports such as airports, railway stations and the like, and important places such as banks, communities and the like closely related to people's life, and the applications have higher requirements on the accuracy and the efficiency of a face recognition algorithm. In order to improve the accuracy of face-based tasks such as face recognition, quality evaluation is usually performed on face images, and then the face images with higher evaluation scores are applied to the tasks. The existing face image quality scoring technology mainly comprises the following steps:
1, scoring according to artificial subjective feelings;
2, no reference evaluation is carried out, according to objectively designed statistical values, such as illumination intensity, ambiguity, face angle and the like are extracted first, and then comprehensive indexes are calculated to serve as label values;
and 3, full reference evaluation, wherein the difference between the test image and the reference image is scored: such as various statistical indicators, such as structural similarity, etc.; or directly using the similarity of the feature vectors extracted by the deep learning network.
However, the above techniques all have disadvantages:
1, the manual scoring cost is high, and the efficiency is low;
2, objective design indexes hardly consider all possible influence factors on one hand, and comprehensive indexes are poor in robustness and inaccurate in quality evaluation on the other hand;
generally, only the difference or the (intra-class) distance between the reference image and the same type of reference image is considered, and the difference or the (inter-class) distance between the reference image and different types of reference images is not considered, so that the evaluation label is not accurate enough, and further the training effect of a subsequent quality evaluation network or model is not good;
and 4, when the quality evaluation is carried out by using the depth characteristics, the accuracy of the depth network is not considered.
In view of the above, a more comprehensive scoring method with high quality evaluation is needed.
Disclosure of Invention
One or more embodiments of the present specification describe a method and an apparatus for labeling face image quality scores, which can solve the problems of low efficiency, high cost and insufficient comprehensive and accurate evaluation in the prior art.
The technical scheme provided by one or more embodiments of the specification is as follows:
in a first aspect, the present invention provides a method for labeling face image quality scores, comprising:
the face image to be detected in the set and the face image with the external noise are subjected to face recognition network to obtain corresponding face characteristic vectors; wherein, each person at least comprises a reference picture;
calculating an indicative function according to the reference map characteristics of the ith person and the reference map characteristics of other different persons;
calculating a quality value of the kth to-be-tested image of the ith person based on the indicative function;
performing linear transformation on the quality value to obtain a final label value of the kth image to be tested of the ith person;
and performing face quality evaluation model training by using the final label value.
In one possible implementation, the out-of-set faces and the in-set face identities do not overlap.
In one possible implementation, the feature vectors are normalized before the indicative function is calculated.
In one possible implementation, the indicative function is calculated according to the following formula:
wherein,
a reference map feature for the ith person;
reference map features for other different persons; j =1, 2.., N + M-1; n is the number of people to be detected in the set, and M is the number of noise people outside the set; the vector inner product of the reference image feature of the ith person and the reference images of other different persons is recorded as
t is a threshold constant.
In one possible implementation, the quality value q of the kth test image of the ith individual is calculated according to the following formula:
therein, ζ i,j Is an indicative function; the vector inner product of the ith individual reference image feature and the kth image feature to be measured is recorded as
p is a relaxation constant; and s is a scale constant.
In one possible implementation, the quality value q is linearly transformed according to the following formula: q (i, j) = w · Q + b, where w, b are constants.
In a second aspect, the present invention provides a face image quality scoring and labeling device, including:
the processing unit is configured to enable the face image to be detected in the set and the face image with the external noise to pass through a face recognition network to obtain corresponding face characteristic vectors; wherein, each person at least comprises a reference picture;
a first calculation unit configured to calculate an indicative function based on the reference map feature of the ith person and the reference map features of the different persons;
a second calculation unit configured to calculate a quality value of the kth chart to be tested of the ith individual based on the indicative function;
the linear transformation unit is configured to perform linear transformation on the quality value to obtain a final label value of the kth to-be-tested graph of the ith person;
and the model training unit is configured to perform face quality evaluation model training by using the final label value.
In a third aspect, the invention provides a face image quality scoring and labeling system, which comprises at least one processor and a memory;
the memory to store one or more program instructions;
the processor is configured to execute one or more program instructions to perform the method according to one or more of the first aspects.
In a fourth aspect, the present invention provides a chip, which is coupled to a memory in a system, so that the chip calls program instructions stored in the memory when running to implement the method according to one or more of the first aspects.
In a fifth aspect, the invention provides a computer readable storage medium comprising one or more program instructions executable by a system according to the third aspect to implement a method according to one or more of the first aspects.
The method provided by the embodiment of the invention comprehensively considers the intra-class distance and the inter-class distance of the human face characteristics and carries out the task of labeling the human face image training sample for the quality evaluation task. By applying the method, more accurate face image quality label values can be obtained, and a foundation is laid for better face quality evaluation networks for subsequent training.
Drawings
Fig. 1 is a schematic flow chart of a method for marking a face image quality score according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a face image quality scoring and labeling device according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a face image quality scoring and labeling system according to an embodiment of the present invention.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not to be construed as limiting the invention. It should be further noted that, for the convenience of description, only the portions related to the related invention are shown in the drawings.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
Fig. 1 shows a flowchart of a method for labeling face image quality score according to an embodiment, where an execution subject of the method may be any device, equipment, platform, or equipment cluster with computing and processing capabilities.
As shown in fig. 1, the method comprises the steps of:
Firstly, establishing an intra-set to-be-detected face image sample set and an external noise face image sample set, specifically, the following procedures are carried out:
the image is processed by a face detection network to obtain a reference image sample set containing N different people
And a sample book to be tested
Each person comprises at least one reference picture;
the images are processed by a face detection network to obtain a reference image sample set containing M different people
Common out-of-collection and in-collection face identities do not overlap
And step 20, calculating the representative function according to the reference picture characteristics of the ith person and the reference picture characteristics of other different persons.
Before calculating the indicative function, normalizing the feature vector, specifically: all pictures are subjected to a face recognition network to obtain corresponding face feature vectors v, and the face feature vectors v are subjected to feature normalization
The reference picture of the ith person is characterized in that
The kth image to be measured of the ith person is characterized in that
Specifically, the illustrative function is calculated by the following formula:
wherein,
a reference map feature for the ith person;
reference map features for other different persons; j =1,2,. Said, N + M-1; n is the number of people to be tested in the set, and M is the number of noise people outside the set; the vector inner product of the reference image feature of the ith person and the reference images of other different persons is recorded as
t is a threshold constant, which may be 0.5.
And step 30, calculating the quality value of the k-th image to be tested of the ith individual based on the indicative function.
Specifically, the quality value q of the kth image to be tested of the ith individual is calculated according to the following formula:
therein, ζ i,j Is an indicative function; the vector inner product of the ith individual's reference image feature and its kth chart feature to be measured is recorded as
p is a relaxation constant, which can be set to 0.2; s is a scale constant, which can be set to 32 in general.
And step 40, performing linear transformation on the quality value to obtain a final label value of the kth image to be tested of the ith individual.
Specifically, the quality value q is linearly transformed according to the following formula: q (i, j) = w · Q + b, where w, b are constants.
And step 50, training a face quality evaluation model by using the final label value.
And (4) performing face quality evaluation model training by using the final label value obtained in the step (40).
The label value is obtained by the features extracted by the deep learning network, so the face image quality scoring and labeling method provided by the embodiment of the invention has high efficiency and low cost. Meanwhile, the method has the following advantages:
(1) The introduction of a relaxation variable is controlled by constructing an indicative function, so that the problem of low intra-class similarity caused by insufficient classification capability of the face recognition model can be solved, and the condition of low score caused by non-picture quality problem is corrected.
(2) By retraining the network model, the face recognition model features are used directly.
(3) Meanwhile, inter-class and intra-class distance influences of the face image are considered, so that evaluation is more comprehensive.
Corresponding to the method of the embodiment, the invention also provides a method and a device for marking the face image quality score, as shown in fig. 2, the method and the device for marking the face image quality score comprise the following steps: a processing unit 210, a first computing unit 220, a second computing unit 230, a linear transformation unit 240, and a model training unit 250. In particular, the method comprises the following steps of,
the processing unit 210 is configured to pass the face image to be detected in the set and the face image with the noise outside the set through a face recognition network to obtain corresponding face feature vectors; wherein, each person at least comprises a reference picture;
a first calculating unit 220 configured to calculate an indicative function based on the reference map feature of the ith person and the reference map features of the different persons;
a second calculation unit 230 configured to calculate a quality value of the kth chart of interest of the ith individual based on the indicative function;
a linear transformation unit 240 configured to perform linear transformation on the quality value to obtain a final label value of the kth to-be-tested image of the ith person;
a model training unit 250 configured to perform face quality assessment model training using the final label value
The functions executed by each component in the face image quality scoring and labeling device provided by the embodiment of the invention are described in detail in the method, so that redundant description is not repeated here.
Corresponding to the above embodiments, the embodiment of the present invention further provides a facial image quality scoring and labeling system, specifically as shown in fig. 3, the system includes at least one processor 310 and a memory 220;
a memory 310 for storing one or more program instructions;
the processor 320 is configured to execute one or more program instructions to perform any method step in the method for labeling face image quality scores as described in the above embodiments.
Corresponding to the above embodiment, an embodiment of the present invention further provides a chip, where the chip is coupled with the memory in the system, so that the chip calls the program instruction stored in the memory when running, thereby implementing the method for scoring and labeling the face image quality introduced in the above embodiment.
Corresponding to the above embodiment, the embodiment of the present invention further provides a computer storage medium, where the computer storage medium includes one or more programs, where one or more program instructions are used to execute the above-described method for labeling face image quality score by a speech recognition system.
The method for marking the face image quality score comprehensively considers the intra-class distance and the inter-class distance of the face features and carries out a task of labeling face image training samples for a quality evaluation task. By applying the method, more accurate face image quality label values can be obtained, and a foundation is laid for better face quality evaluation networks for subsequent training. The label value is obtained by the features extracted by the deep learning network, so the method has high efficiency and low cost, and simultaneously considers the inter-class and intra-class distance influence of the face image, thereby ensuring more comprehensive evaluation.
Those of skill would further appreciate that the various illustrative components and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the components and steps of the various examples have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied in hardware, a software module executed by a processor, or a combination of the two. A software module may reside in Random Access Memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The above embodiments, objects, technical solutions and advantages of the present invention are further described in detail, it should be understood that the above embodiments are only examples of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (7)
1. A face image quality scoring and labeling method is characterized by comprising the following steps:
the face image to be detected in the set and the face image with the external noise are subjected to face recognition network to obtain corresponding face characteristic vectors; wherein, each person at least comprises a reference picture;
calculating an indicative function according to the reference graph characteristics of the ith person and the reference graph characteristics of other different persons;
calculating a quality value for the kth map of the ith individual based on the indicative function;
performing linear transformation on the quality value to obtain a final label value of the kth image to be tested of the ith person;
performing face quality evaluation model training by using the final label value;
an exemplary function is calculated according to the following formula:
wherein,
a reference map feature for the ith person;
reference map features for different other persons; j =1,2,. Said, N + M-1; n is the number of people to be tested in the set, and M is the number of noise people outside the set; the vector inner product of the reference image feature of the ith person and the reference images of other different persons is recorded as
t is a threshold constant;
calculating the quality value q of the kth image to be tested of the ith person according to the following formula:
therein, ζ i，j Is an indicative function; the vector inner product of the ith individual's reference image feature and its kth chart feature to be measured is recorded as
p is a relaxation constant; s is a scale constant.
2. The method of claim 1, wherein the out-of-collection face and the in-collection face identities do not overlap.
3. The method of claim 1, wherein the feature vectors are normalized prior to computing the indicative function.
4. The method of claim 1, wherein the quality value q is linearly transformed according to the following equation: q (i, j) = w · Q + b, where w, b are constants.
5. A face image quality scoring and labeling device is characterized by comprising:
the processing unit is configured to enable the face image to be detected in the set and the face image with the external noise to pass through a face recognition network to obtain corresponding face characteristic vectors; wherein, each person at least comprises a reference picture;
a first calculation unit configured to calculate an indicative function based on the reference map feature of the ith person and the reference map features of the different persons;
a second calculation unit configured to calculate a quality value of the kth chart to be tested of the ith person based on the indicative function;
the linear transformation unit is configured to perform linear transformation on the quality value to obtain a final label value of the kth image to be tested of the ith person;
the model training unit is configured to perform face quality evaluation model training by using the final label value;
an illustrative function is calculated according to the following formula:
wherein,
a reference map feature for the ith person;
reference map features for different other persons; j =1,2,. Said, N + M-1; n is the number of people to be detected in the set, and M is the number of noise people outside the set; the vector inner product of the reference image feature of the ith person and the reference images of other different persons is recorded as
t is a threshold constant;
calculating the quality value q of the kth image to be tested of the ith person according to the following formula:
therein, ζ i，j Is an indicative function; the vector inner product of the ith individual's reference image feature and its kth chart feature to be measured is recorded as
p is a relaxation constant; s is a scale constant.
6. A face image quality scoring and labeling system is characterized by comprising at least one processor and a memory;
the memory to store one or more program instructions;
the processor operable to execute one or more program instructions to perform the method of any of claims 1 to 4.
7. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010822252.8A CN111967381B (en) | 2020-08-16 | 2020-08-16 | Face image quality grading and labeling method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010822252.8A CN111967381B (en) | 2020-08-16 | 2020-08-16 | Face image quality grading and labeling method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111967381A CN111967381A (en) | 2020-11-20 |
| CN111967381B true CN111967381B (en) | 2022-11-11 |
Family
ID=73388178
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010822252.8A Active CN111967381B (en) | 2020-08-16 | 2020-08-16 | Face image quality grading and labeling method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111967381B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113139462A (en) * | 2021-04-23 | 2021-07-20 | 杭州魔点科技有限公司 | Unsupervised face image quality evaluation method, electronic device and storage medium |
| CN113192028B (en) * | 2021-04-29 | 2022-05-31 | 合肥的卢深视科技有限公司 | Quality evaluation method and device for face image, electronic equipment and storage medium |
| CN113688896A (en) * | 2021-08-20 | 2021-11-23 | 杭州英歌智达科技有限公司 | Face image quality labeling method and device based on face recognition system |
| CN114549409A (en) * | 2022-01-11 | 2022-05-27 | 深圳市联洲国际技术有限公司 | Image processing method, image processing device, storage medium and computer equipment |
| CN117372405A (en) * | 2023-10-31 | 2024-01-09 | 神州通立电梯有限公司 | Face image quality evaluation method, device, storage medium and equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107330397A (en) * | 2017-06-28 | 2017-11-07 | 苏州经贸职业技术学院 | A kind of pedestrian's recognition methods again based on large-spacing relative distance metric learning |
| CN108235770A (en) * | 2017-12-29 | 2018-06-29 | 深圳前海达闼云端智能科技有限公司 | image identification method and cloud system |
| CN109003266A (en) * | 2018-07-13 | 2018-12-14 | 中国科学院长春光学精密机械与物理研究所 | A method of based on fuzzy clustering statistical picture quality subjective evaluation result |
| CN109241880A (en) * | 2018-08-22 | 2019-01-18 | 北京旷视科技有限公司 | Image processing method, image processing apparatus, computer readable storage medium |
| CN111241925A (en) * | 2019-12-30 | 2020-06-05 | 新大陆数字技术股份有限公司 | Face quality evaluation method, system, electronic equipment and readable storage medium |
-
2020
- 2020-08-16 CN CN202010822252.8A patent/CN111967381B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107330397A (en) * | 2017-06-28 | 2017-11-07 | 苏州经贸职业技术学院 | A kind of pedestrian's recognition methods again based on large-spacing relative distance metric learning |
| CN108235770A (en) * | 2017-12-29 | 2018-06-29 | 深圳前海达闼云端智能科技有限公司 | image identification method and cloud system |
| CN109003266A (en) * | 2018-07-13 | 2018-12-14 | 中国科学院长春光学精密机械与物理研究所 | A method of based on fuzzy clustering statistical picture quality subjective evaluation result |
| CN109241880A (en) * | 2018-08-22 | 2019-01-18 | 北京旷视科技有限公司 | Image processing method, image processing apparatus, computer readable storage medium |
| CN111241925A (en) * | 2019-12-30 | 2020-06-05 | 新大陆数字技术股份有限公司 | Face quality evaluation method, system, electronic equipment and readable storage medium |
Non-Patent Citations (2)
| Title |
|---|
| Face Quality Estimation and Its Correlation to Demographic and Non-Demographic Bias in Face Recognition;Philipp Terhorst等;《arXiv:2004.01019v3[cs.CV]》;20200710;全文 * |
| Philipp Terhörst等.SER-FIQ: Unsupervised Estimation of Face Image Quality Based on Stochastic Embedding Robustness.《2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)》.2020, * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111967381A (en) | 2020-11-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111967381B (en) | Face image quality grading and labeling method and device | |
| US11610394B2 (en) | Neural network model training method and apparatus, living body detecting method and apparatus, device and storage medium | |
| CN109065027B (en) | Voice distinguishing model training method and device, computer equipment and storage medium | |
| WO2020211339A1 (en) | Finger vein recognition method and apparatus, and computer device and storage medium | |
| US7873189B2 (en) | Face recognition by dividing an image and evaluating a similarity vector with a support vector machine | |
| EP3696566A1 (en) | System and method for identification of an airborne object | |
| CN109614881B (en) | Biometric authentication method, device and storage device capable of adaptively adjusting threshold | |
| CN104123543B (en) | A kind of eye movement recognition methods based on recognition of face | |
| CN108304820A (en) | A kind of method for detecting human face, device and terminal device | |
| CN114722812B (en) | Multi-mode deep learning model vulnerability analysis method and system | |
| CN104680545B (en) | There is the detection method of well-marked target in optical imagery | |
| CN111178252A (en) | Multi-feature fusion identity recognition method | |
| CN114241505B (en) | Method and device for extracting chemical structure image, storage medium and electronic equipment | |
| Morrison | Vowel inherent spectral change in forensic voice comparison | |
| CN110874576B (en) | Pedestrian re-identification method based on typical correlation analysis fusion characteristics | |
| CN110909889A (en) | Training set generation and model training method and device based on feature distribution | |
| CN111785262A (en) | A Speaker Age and Gender Classification Method Based on Residual Network and Fusion Features | |
| CN110766657B (en) | A method for evaluating the quality of laser interference images | |
| US20190205589A1 (en) | Latent fingerprint ridge flow map improvement | |
| RU2148274C1 (en) | Method for identification of person using properties of signature | |
| CN118097755A (en) | Intelligent face identity recognition method based on YOLO network | |
| CN116153337B (en) | Synthetic voice tracing evidence obtaining method and device, electronic equipment and storage medium | |
| CN110909678A (en) | A face recognition method and system based on width learning network feature extraction | |
| CN117727084A (en) | Face recognition system and method based on big data | |
| CN113436735A (en) | Body weight index prediction method, device and storage medium based on face structure measurement |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |