CN110580461A - Facial expression recognition algorithm combined with multilevel convolution characteristic pyramid - Google Patents
Facial expression recognition algorithm combined with multilevel convolution characteristic pyramid Download PDFInfo
- Publication number
- CN110580461A CN110580461A CN201910806700.2A CN201910806700A CN110580461A CN 110580461 A CN110580461 A CN 110580461A CN 201910806700 A CN201910806700 A CN 201910806700A CN 110580461 A CN110580461 A CN 110580461A
- Authority
- CN
- China
- Prior art keywords
- network
- feature
- facial expression
- feature extraction
- level
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
-
- 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
- G06V40/174—Facial expression recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a facial expression recognition algorithm combined with a multilevel convolution characteristic pyramid, which comprises the following steps: inputting the original facial expression picture into a first feature extraction network for feature extraction, and extracting global features; inputting the cut and amplified facial expression picture into a second feature extraction network for feature extraction, and extracting local features; constructing a feature pyramid network in the first feature extraction network and the second feature extraction network; carrying out regional positioning on the local features in the second feature extraction network by using an attention regional positioning network; and fusing the global features and the local features of the human face expression pictures by using the feature fusion network, and classifying the human face expression pictures through a full connection layer. The aim of improving the accuracy of the facial expression recognition technology on the facial expression recognition is achieved.
Description
Technical Field
the invention relates to the field of image processing, in particular to a facial expression recognition algorithm combined with a multilevel convolution characteristic pyramid.
Background
The human face expression is not only a kind of emotion of people, but also can transmit abundant human behavior information. The facial expression recognition technology is a main form for researching human emotion, and the accurate recognition of the facial expression can effectively judge the emotional state of a human. The facial expression recognition can be applied to many fields, such as the field of safe driving, the field of intelligent human-computer interaction, the field of medical monitoring, lie detection in criminal investigation, psychotherapy and the like. However, the accuracy of the existing facial expression recognition technology for facial expression recognition is low.
Disclosure of Invention
the invention aims to provide a facial expression recognition algorithm combined with a multilevel convolution characteristic pyramid, and aims to solve the technical problem that the facial expression recognition technology in the prior art is low in accuracy rate of facial expression recognition.
In order to achieve the above object, the facial expression recognition algorithm of the invention, which combines a multilevel convolution feature pyramid, comprises the following steps:
Inputting the original facial expression picture into a first feature extraction network for feature extraction, and extracting global features;
inputting the cut and amplified facial expression picture into a second feature extraction network for feature extraction, and extracting local features;
Constructing a feature pyramid network in the first feature extraction network and the second feature extraction network;
carrying out regional positioning on the local features in the second feature extraction network by using an attention regional positioning network;
and fusing the global features and the local features of the human face expression pictures by using the feature fusion network, and classifying the human face expression pictures through a full connection layer.
The input sample of the second-level feature extraction network is obtained by cutting and amplifying an input sample picture in the first-level feature extraction network.
The first-level feature extraction network and the second-level feature extraction network are both full convolution structure networks.
Wherein, the convolution kernel with the size of 3 x 3 is used in the first convolution layer of the global feature extraction and the local feature extraction.
the method comprises the following steps of inputting an original facial expression picture into a first-level feature extraction network, extracting features of the facial expression picture through a convolutional neural network, and then outputting categories and probability values of each category, wherein the probability value of each category is obtained by the following method:
C(i)＝f(Wi⊙Xi)
p(i)＝f(Wi⊙Xi)
Wherein, the input facial expression picture is X, Xiand Wirepresents the input and weight parameters of the i-th convolutional neural network, indicates the convolution, pooling, activation, or normalization operation performed during convolution, f (-) indicates the feature extraction network, C(i)(i ═ 1,2) denotes an output class label of the i-th convolutional neural network, p(i)(i ═ 1,2) denotes the probability value of the i-th order convolutional neural network for outputting each class.
Wherein, each level of network generates a feature map about the input facial expression picture for userepresenting a set of generated feature maps of each stage of the network, wherein m represents the number of network stages contained in the model, wherein m is equal to (1,2), and n represents the final output of each stage of the networkthe input image X is represented by the following sub-formula:
the feature graph generated by the convolutional layer at the later layer is transposed and convolved, and then the two feature graphs are added to unify the scales of the feature graphs.
Wherein the attention area positioning network maps the input feature map into (D)x,Dy) Is a center, DαThe attention area is a square attention area with side length, then the attention area is mapped to an original facial expression picture, and the original facial expression picture is cut and amplified and then sent to a next-level network.
the second feature extraction network can automatically activate the salient region of the facial expression during recognition and extraction of the facial expression, and firstly, accurate positioning must be realized on the region with the largest response in the feature map, and then the accurately positioned regions are mapped to the original facial expression picture to obtain the important local region of the original facial expression, and feature extraction is carried out on the local region to realize local region recognition of the picture.
When initializing, all feature maps generated by the last layer of the first-level feature extraction network are added together, then an area with a larger numerical value in the added feature maps is positioned, the positioned area is fitted into a square, then the central coordinate and the side length of the square are obtained and set as initialization parameters of the attention network, and the initialization parameters are expressed as:
Wherein F represents a feature map generated by the last convolutional layer of the first feature extraction network, d represents the total number of feature maps generated by the network, F is the total feature map obtained by adding each corresponding point of each feature map, h and w represent the height and width of the feature map, and p represents the pixel mean value of the total feature map.
the invention relates to a facial expression recognition algorithm combined with a multilevel convolution feature pyramid, which is characterized in that an original facial expression picture is input into a first feature extraction network for feature extraction, and global features are extracted; inputting the cut and amplified facial expression picture into a second feature extraction network for feature extraction, and extracting local features; constructing a feature pyramid network in the first feature extraction network and the second feature extraction network; carrying out regional positioning on the local features in the second feature extraction network by using an attention regional positioning network; and fusing the global features and the local features of the human face expression pictures by using the feature fusion network, and classifying the human face expression pictures through a full connection layer. The human face features are transferred from the whole to the local by adopting a cascade structure of two levels of human face feature extraction networks, and a feature pyramid network is added into each level of human face feature extraction network, so that the robustness of the algorithm to the size of the human face target is improved. Therefore, the effect of improving the accuracy of the facial expression recognition technology on the facial expression recognition is obtained.
drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a step diagram of the facial expression recognition algorithm of the present invention incorporating a multi-level convolution feature pyramid.
FIG. 2 is a flow chart of the facial expression recognition algorithm of the present invention incorporating a multi-level convolution feature pyramid.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
in the description of the present invention, it is to be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on the orientations or positional relationships illustrated in the drawings, and are used merely for convenience in describing the present invention and for simplicity in description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed in a particular orientation, and be operated, and thus, are not to be construed as limiting the present invention. Further, in the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.
referring to fig. 1 and fig. 2, the present invention provides a facial expression recognition algorithm combining a multilevel convolution feature pyramid, which includes the following steps:
S100: inputting the original facial expression picture into a first feature extraction network for feature extraction, and extracting global features;
S200: inputting the cut and amplified facial expression picture into a second feature extraction network for feature extraction, and extracting local features;
s300: constructing a feature pyramid network in the first feature extraction network and the second feature extraction network;
S400: carrying out regional positioning on the local features in the second feature extraction network by using an attention regional positioning network;
s500: and fusing the global features and the local features of the human face expression pictures by using the feature fusion network, and classifying the human face expression pictures through a full connection layer.
In this embodiment, the first-level feature extraction network can perform feature extraction on 448 × 448 pixels of an original facial expression picture, and extract global features; the second-level feature extraction network is used for extracting features of the cut and amplified human face expression picture and extracting local features; the input sample of the second-level feature extraction network is obtained by cutting and amplifying an input sample picture in the first-level feature extraction network, and the first feature extraction network and the second feature extraction network are both full convolution structure networks. The structure of the network is similar to that of the VGG19 network, each module of the network is composed of a convolution layer, a BatchNorm layer, a relu layer and an average pooling layer, and 16 convolution modules are provided. The first convolutional layer in both global feature extraction and local feature extraction uses a convolution kernel of size 3 x 3.
inputting the original facial expression picture into a first-level feature extraction network, extracting the features of the facial expression picture through a convolutional neural network, and then outputting the category and the probability value of each category, which are expressed by the following sub-expressions:
C(i)＝f(Wi⊙Xi)
p(i)＝f(Wi⊙Xi)
In the above formula, the input facial expression picture is X, XiAnd Wirepresents the input and weight parameters of the i-th convolutional neural network, indicates the convolution, pooling, activation, or normalization operation performed during convolution, f (-) indicates the feature extraction network, C(i)(i ═ 1,2) denotes an output class label of the i-th convolutional neural network, p(i)(i ═ 1,2) denotes the probability value of the i-th order convolutional neural network for outputting each class.
Each level of network generates a feature map about the input facial expression pictureRepresenting that each stage of network generates a set of feature maps, m represents the number of network stages contained in the model, wherein m is (1,2), n represents the number of the feature maps finally output by each stage of network, and the input image X is represented by the following formula:
The convolutional neural network extracts features of an input picture layer by layer through convolutional cores of each layer, feature graphs of different convolutional layers correspond to different regions and targets of the input picture, in order to solve the problem of face size diversification, and in order to better utilize shallow convolutional layers of the convolutional neural network to extract face expression information, a feature pyramid network is constructed among convolutional layers of the convolutional neural network, the feature graph scale generated by the shallow convolutional layers in the convolutional neural network is large, the feature graph scale generated by the convolutional layers at the body of the convolutional neural network is small, in order to unify the scale of the feature graphs, firstly, feature graphs generated by the convolutional layers of the next layer are subjected to transposition convolution operation, then, the two feature graphs are added, and the generated feature graph is the final representation of the whole face expression picture.
in order to realize transition and transfer of facial expressions from global features to local features, the attention area positioning network is arranged between feature extraction networks and is responsible for positioning local feature areas (such as eyebrows, eyes, a nose and a mouth) in an original facial expression picture, cutting the corresponding areas and sending the areas to a next-level network. The trained facial expression recognition network can automatically activate the salient region of the facial expression (namely, the region with important influence on the facial expression recognition effect) when the facial expression is recognized, so that the region with the largest response in the feature map must be accurately positioned, and then the accurately positioned regions are mapped to the original facial expression picture, so that the very important local region of the original facial expression can be obtained, and the local region is subjected to feature extraction to realize the local region recognition of the picture. The regions having important influence on facial expression recognition are automatically located by adopting the step-by-step location attention network.
the attention area positioning network maps the input feature map into (D)x，Dy) Is a center, DaThe attention area is a square attention area with side length, the attention area is mapped to an original facial expression picture, and the original facial expression picture is cut and amplified and then sent to a next-level network, so that the facial expression picture can be finely identified from a global feature picture to a local feature area.
the key point of mapping the strong response area in the feature map obtained through the convolutional neural network to the original facial expression picture is to obtain the coordinates of the strong response area in the feature map, and then the feature map can be automatically positioned through the full-connection layer network by designing an attention area positioning network of a 3-layer full-connection neural network. In order to accelerate the convergence speed of the attention area positioning network, all feature maps generated by the last layer of the first-stage feature extraction network are added together during initialization, then an area with a larger numerical value in the added feature maps is positioned, the positioned area is fitted into a square, then the central coordinate and the side length of the square are obtained and set as initialization parameters of the attention area positioning network, and the specific implementation mode is as follows:
in the above formula, F represents a feature map generated by the last convolutional layer of the first feature extraction network, d represents the total number of feature maps generated by the network, F is the total feature map obtained by adding each corresponding point of each feature map, h and w represent the height and width of the feature map, p represents the pixel mean value of the total feature map, then each pixel point of the total feature map is compared with the pixel mean value, which is equivalent to the threshold value taking the pixel mean value as the total feature map, the pixel points larger than the threshold value are set to be 1, the pixel points smaller than or equal to the threshold value are set to be 0, then the longest edge of the maximum connected region set to be 1 is selected to be the edge length of the square, and finally the central coordinate and the edge length of the square are initialized to the parameters of the attention localization network. The attention area positioning network can automatically position the area with the largest response in the feature map, and then cuts and enlarges the attention area after positioning, wherein the attention area is supposed to be gradually increased from left to right and from bottom to top by taking the lower left corner of the picture as the origin and the horizontal direction as the x axis. According to the obtained coordinates of the center point of the square attention area and the side length of the square attention area, four vertexes of the square attention area can be calculated, so that automatic clipping of the attention area is achieved, and finally the attention area picture is amplified by utilizing a bilinear interpolation value. To obtain the picture we need.
Each level of feature extraction network of the algorithm can extract features of the facial expression picture, the first feature extraction network extracts global features of the facial expression picture, the second network extracts local detail features of the facial expression picture, pixel points corresponding to feature pictures of each channel are added, feature dimension reduction is conducted through convolution with the size of 1 x 1, then feature fusion and pooling are conducted, and finally category labels of the facial expression are output through a full connection layer.
After the network is designed according to the method, in order to simultaneously optimize the feature extraction network and the attention area positioning network, two loss functions are adopted to optimize the network, the weighting sum of the cross entropy loss functions of the two networks and the cross entropy loss function of the fusion network is used as the integral loss function of the network, so that the effect of distinguishing different facial expressions is good, and meanwhile, in order to restrict the difference of similar facial expressions, the punishment verification loss of the two networks of the feature extraction network and the attention area positioning network is adopted, so that the recognition performance between similar facial expressions is greatly enhanced.
In summary, the following steps: the algorithm can automatically position the area with larger distinguishing degree of the facial expression picture, and the global and local characteristics of the facial expression picture are fused, so that the accurate recognition of the facial expression picture is realized. The first feature extraction network and the second feature extraction network are responsible for carrying out feature extraction on the facial expression pictures, the attention area positioning network is responsible for positioning the local feature areas of the facial expression pictures, expressing and positioning the local feature areas of the facial expression pictures, and for the facial expression pictures with high similarity, such as anger and disgust, the algorithm realizes the area positioning of the local features of the facial expression with high discrimination, and comprehensively utilizes the superficial layer high-resolution face structure information and the deep layer low-resolution semantic information of the convolutional neural network to obtain better positioning of the local feature areas of the facial expression, so that the recognition rate of the facial expression pictures with high similarity is greatly improved.
The feature fusion network is responsible for fusing the global features and the local features of the facial expressions and fusing the global feature information of the facial expression pictures and the local feature information of the facial expression pictures. The local feature information of the facial expression pictures can improve the distinguishing capability of facial expressions with larger similarity, and meanwhile, the global structure information of the facial expression pictures is kept, so that the features extracted by all levels of facial expression feature extraction networks are fused by adopting a feature fusion network, more comprehensive feature information of the facial expression pictures is obtained, and the accuracy of facial expression recognition is improved. Because the single-pole facial expression feature extraction network cannot simultaneously obtain the global and local features of the facial expression pictures, the two-stage facial expression feature extraction network cascade structure is adopted to realize the transfer of the features from the global to the local, and a feature pyramid network is constructed between the feature images of each stage of facial expression feature extraction network aiming at the characteristic that the facial expression pictures have scale changes, so that the feature description capacity of the network is improved, and the identification accuracy of the facial expression pictures with scale changes, the facial expression pictures with shelters and the facial expression pictures with similarity is greatly improved.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. A facial expression recognition algorithm combined with a multilevel convolution feature pyramid is characterized by comprising the following steps:
Inputting the original facial expression picture into a first feature extraction network for feature extraction, and extracting global features;
Inputting the cut and amplified facial expression picture into a second feature extraction network for feature extraction, and extracting local features;
Constructing a feature pyramid network in the first feature extraction network and the second feature extraction network;
Carrying out regional positioning on the local features in the second feature extraction network by using an attention regional positioning network;
And fusing the global features and the local features of the human face expression pictures by using the feature fusion network, and classifying the human face expression pictures through a full connection layer.
2. The algorithm for facial expression recognition in combination with the multi-level pyramid of convolution features of claim 1,
The input sample of the second-level feature extraction network is obtained by cutting and amplifying an input sample picture in the first-level feature extraction network.
3. The algorithm for facial expression recognition in combination with the multi-level pyramid of convolution features of claim 1,
The first level feature extraction network and the second level feature extraction network are both full convolution structure networks.
4. The algorithm for facial expression recognition in combination with the multi-level pyramid of convolution features of claim 3,
The first convolutional layer in both global feature extraction and local feature extraction uses a convolution kernel of size 3 x 3.
5. The algorithm for facial expression recognition in combination with the pyramid of multilevel convolution features of claim 4,
Inputting an original facial expression picture into a first-level feature extraction network, extracting features of the facial expression picture through a convolutional neural network, and then outputting categories and probability values of each category, wherein the probability value of each category is obtained by the following steps:
C(i)＝f(Wi⊙Xi)
p(i)＝f(Wi⊙Xi)
Wherein, the input facial expression picture is X, XiAnd Wirepresents the input and weight parameters of the i-th convolutional neural network, indicates the convolution, pooling, activation, or normalization operation performed during convolution, f (-) indicates the feature extraction network, C(i)(i ═ 1,2) denotes an output class label of the i-th convolutional neural network, p(i)(i ═ 1,2) denotes the probability value of the i-th order convolutional neural network for outputting each class.
6. The algorithm for facial expression recognition in combination with the pyramid of multilevel convolution features of claim 5,
Each level of network generates a feature map about the input facial expression picturerepresenting a set of generated feature maps of each level of network, wherein m represents the number of network levels contained in the model, wherein m is equal to (1,2), n represents the number of the feature maps finally output by each level of network, and an input image X is represented as:
7. The algorithm for facial expression recognition in combination with the multi-level pyramid of convolution features of claim 6,
The feature graph generated by the shallow convolutional layer in the convolutional neural network is large in scale, the feature graph generated by the deep convolutional layer in the convolutional neural network is small in scale, the feature graph generated by the next convolutional layer is transposed and convolved, and then the two feature graphs are added to unify the scales of the feature graphs.
8. the algorithm for facial expression recognition in combination with the multi-level pyramid of convolution features of claim 7,
the attention area positioning network maps the input feature map into (D)x,Dy) Is a center, Dαthe attention area is a square attention area with side length, then the attention area is mapped to an original facial expression picture, and the original facial expression picture is cut and amplified and then sent to a next-level network.
9. The algorithm for facial expression recognition in combination with the pyramid of multilevel convolution features of claim 8,
The second feature extraction network can automatically activate the salient regions of the facial expressions in the process of identifying and extracting the facial expressions, firstly, accurate positioning must be achieved on the regions with the largest response in the feature images, then the accurately positioned regions are mapped to the original facial expression images, important local regions of the original facial expressions are obtained, and feature extraction is conducted on the local regions to achieve local region identification of the images.
10. The algorithm for facial expression recognition in combination with the multi-level pyramid of convolution features of claim 9,
When initializing, all feature maps generated by the last layer of the first-level feature extraction network are added together, then an area with a larger numerical value in the added feature maps is positioned, the positioned area is fitted into a square, then the central coordinate and the side length of the square are obtained and set as initialization parameters of the attention network, and the initialization parameters are expressed as:
Wherein F represents a feature map generated by the last convolutional layer of the first feature extraction network, d represents the total number of feature maps generated by the network, F is the total feature map obtained by adding each corresponding point of each feature map, h and w represent the height and width of the feature map, and p represents the pixel mean value of the total feature map.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910806700.2A CN110580461A (en) | 2019-08-29 | 2019-08-29 | Facial expression recognition algorithm combined with multilevel convolution characteristic pyramid |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910806700.2A CN110580461A (en) | 2019-08-29 | 2019-08-29 | Facial expression recognition algorithm combined with multilevel convolution characteristic pyramid |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110580461A true CN110580461A (en) | 2019-12-17 |
Family
ID=68812404
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910806700.2A Pending CN110580461A (en) | 2019-08-29 | 2019-08-29 | Facial expression recognition algorithm combined with multilevel convolution characteristic pyramid |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110580461A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111368663A (en) * | 2020-02-25 | 2020-07-03 | 华南理工大学 | Method, device, medium and equipment for recognizing static facial expressions in natural scene |
| CN112036260A (en) * | 2020-08-10 | 2020-12-04 | 武汉星未来教育科技有限公司 | Expression recognition method and system for multi-scale sub-block aggregation in natural environment |
| CN112200065A (en) * | 2020-10-09 | 2021-01-08 | 福州大学 | Micro-expression classification method based on action amplification and self-adaptive attention area selection |
| CN112364827A (en) * | 2020-11-30 | 2021-02-12 | 腾讯科技（深圳）有限公司 | Face recognition method and device, computer equipment and storage medium |
| CN112464865A (en) * | 2020-12-08 | 2021-03-09 | 北京理工大学 | Facial expression recognition method based on pixel and geometric mixed features |
| CN112800914A (en) * | 2021-01-20 | 2021-05-14 | 桂林电子科技大学 | Face recognition method based on expert prior knowledge and LTP complex illumination |
| CN113361493A (en) * | 2021-07-21 | 2021-09-07 | 天津大学 | Facial expression recognition method for robustness of different image resolutions |
| CN113569732A (en) * | 2021-07-27 | 2021-10-29 | 厦门理工学院 | Face attribute recognition method and system based on parallel sharing multitask network |
| CN115205986A (en) * | 2022-08-09 | 2022-10-18 | 山东省人工智能研究院 | False video detection method based on knowledge distillation and transformer |
| WO2023029678A1 (en) * | 2022-04-06 | 2023-03-09 | 江苏商贸职业学院 | Gis-based agricultural service management method and system |
| CN115909455A (en) * | 2022-11-16 | 2023-04-04 | 航天恒星科技有限公司 | Expression recognition method integrating multi-scale feature extraction and attention mechanism |
| WO2023173646A1 (en) * | 2022-03-17 | 2023-09-21 | 深圳须弥云图空间科技有限公司 | Expression recognition method and apparatus |
| WO2024001095A1 (en) * | 2022-06-27 | 2024-01-04 | 闻泰通讯股份有限公司 | Facial expression recognition method, terminal device and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409222A (en) * | 2018-09-20 | 2019-03-01 | 中国地质大学（武汉） | A kind of multi-angle of view facial expression recognizing method based on mobile terminal |
| CN109815924A (en) * | 2019-01-29 | 2019-05-28 | 成都旷视金智科技有限公司 | Expression recognition method, apparatus and system |
-
2019
- 2019-08-29 CN CN201910806700.2A patent/CN110580461A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409222A (en) * | 2018-09-20 | 2019-03-01 | 中国地质大学（武汉） | A kind of multi-angle of view facial expression recognizing method based on mobile terminal |
| CN109815924A (en) * | 2019-01-29 | 2019-05-28 | 成都旷视金智科技有限公司 | Expression recognition method, apparatus and system |
Non-Patent Citations (1)
| Title |
|---|
| 梁华刚等: "多级卷积特征金字塔的细粒度食物图片识别", 《中国图象图形学报》 * |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111368663B (en) * | 2020-02-25 | 2024-02-20 | 华南理工大学 | Method, device, medium and equipment for recognizing static facial expression in natural scene |
| CN111368663A (en) * | 2020-02-25 | 2020-07-03 | 华南理工大学 | Method, device, medium and equipment for recognizing static facial expressions in natural scene |
| CN112036260A (en) * | 2020-08-10 | 2020-12-04 | 武汉星未来教育科技有限公司 | Expression recognition method and system for multi-scale sub-block aggregation in natural environment |
| CN112200065B (en) * | 2020-10-09 | 2022-08-09 | 福州大学 | Micro-expression classification method based on action amplification and self-adaptive attention area selection |
| CN112200065A (en) * | 2020-10-09 | 2021-01-08 | 福州大学 | Micro-expression classification method based on action amplification and self-adaptive attention area selection |
| CN112364827A (en) * | 2020-11-30 | 2021-02-12 | 腾讯科技（深圳）有限公司 | Face recognition method and device, computer equipment and storage medium |
| CN112364827B (en) * | 2020-11-30 | 2023-11-10 | 腾讯科技（深圳）有限公司 | Face recognition method, device, computer equipment and storage medium |
| CN112464865A (en) * | 2020-12-08 | 2021-03-09 | 北京理工大学 | Facial expression recognition method based on pixel and geometric mixed features |
| CN112800914A (en) * | 2021-01-20 | 2021-05-14 | 桂林电子科技大学 | Face recognition method based on expert prior knowledge and LTP complex illumination |
| CN113361493B (en) * | 2021-07-21 | 2022-05-20 | 天津大学 | Facial expression recognition method robust to different image resolutions |
| CN113361493A (en) * | 2021-07-21 | 2021-09-07 | 天津大学 | Facial expression recognition method for robustness of different image resolutions |
| CN113569732A (en) * | 2021-07-27 | 2021-10-29 | 厦门理工学院 | Face attribute recognition method and system based on parallel sharing multitask network |
| CN113569732B (en) * | 2021-07-27 | 2023-06-06 | 厦门理工学院 | Face attribute identification method and system based on parallel sharing multitasking network |
| WO2023173646A1 (en) * | 2022-03-17 | 2023-09-21 | 深圳须弥云图空间科技有限公司 | Expression recognition method and apparatus |
| WO2023029678A1 (en) * | 2022-04-06 | 2023-03-09 | 江苏商贸职业学院 | Gis-based agricultural service management method and system |
| WO2024001095A1 (en) * | 2022-06-27 | 2024-01-04 | 闻泰通讯股份有限公司 | Facial expression recognition method, terminal device and storage medium |
| CN115205986A (en) * | 2022-08-09 | 2022-10-18 | 山东省人工智能研究院 | False video detection method based on knowledge distillation and transformer |
| CN115205986B (en) * | 2022-08-09 | 2023-05-19 | 山东省人工智能研究院 | Dummy video detection method based on knowledge distillation and transformation |
| CN115909455B (en) * | 2022-11-16 | 2023-09-19 | 航天恒星科技有限公司 | Expression recognition method integrating multi-scale feature extraction and attention mechanism |
| CN115909455A (en) * | 2022-11-16 | 2023-04-04 | 航天恒星科技有限公司 | Expression recognition method integrating multi-scale feature extraction and attention mechanism |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110580461A (en) | Facial expression recognition algorithm combined with multilevel convolution characteristic pyramid | |
| CN111340814B (en) | RGB-D image semantic segmentation method based on multi-mode self-adaptive convolution | |
| Yang et al. | Towards real-time traffic sign detection and classification | |
| CN111639544B (en) | Expression recognition method based on multi-branch cross-connection convolutional neural network | |
| CN105528575B (en) | Sky detection method based on Context Reasoning | |
| CN112132197B (en) | Model training, image processing method, device, computer equipment and storage medium | |
| CN112801015B (en) | Multi-mode face recognition method based on attention mechanism | |
| CN111639564B (en) | Video pedestrian re-identification method based on multi-attention heterogeneous network | |
| CN112766229B (en) | Human face point cloud image intelligent identification system and method based on attention mechanism | |
| CN112016601B (en) | Network model construction method based on knowledge graph enhanced small sample visual classification | |
| Yang et al. | Facial expression recognition based on dual-feature fusion and improved random forest classifier | |
| CN109165658B (en) | Strong negative sample underwater target detection method based on fast-RCNN | |
| CN115019274B (en) | Road surface disease identification method integrating tracking and searching algorithms | |
| CN112836625A (en) | Face living body detection method and device and electronic equipment | |
| CN112396036B (en) | Method for re-identifying blocked pedestrians by combining space transformation network and multi-scale feature extraction | |
| CN113378675A (en) | Face recognition method for simultaneous detection and feature extraction | |
| CN111368775A (en) | Complex scene dense target detection method based on local context sensing | |
| CN113011253A (en) | Face expression recognition method, device, equipment and storage medium based on ResNeXt network | |
| CN115880397A (en) | Topological semantic map construction method for blind assisting scene | |
| CN110414431B (en) | Face recognition method and system based on elastic context relation loss function | |
| CN114529842A (en) | Figure interaction detection method based on two-way attention mechanism under knowledge guidance | |
| CN111881803B (en) | Face recognition method based on improved YOLOv3 | |
| CN116665265A (en) | Facial living body detection method and system based on facial feature segmentation | |
| CN116311518A (en) | Hierarchical character interaction detection method based on human interaction intention information | |
| CN111898473B (en) | Driver state real-time monitoring method based on deep learning |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191217 |