137 results about "Face aging" patented technology

Age-estimation of a face of an individual is represented in image data. In one embodiment, age-estimation techniques involves combining a Contourlet Appearance Model (CAM) for facial-age feature extraction and Support Vector Regression (SVR) for learning aging rules in order to improve the accuracy of age-estimation over the current techniques. In a particular example, characteristics of input facial images are converted to feature vectors by CAM, then these feature vectors are analyzed by an aging-mechanism-based classifier to estimate whether the images represent faces of younger or older people prior to age-estimation, the aging-mechanism-based classifier being generated in one embodiment by running Support Vector Machines (SVM) on training images. In an exemplary binary youth/adult classifier, faces classified as adults are passed to an adult age-estimation function and the others are passed to a youth age-estimation function.

The invention provides a face automatic aging mechanism based on a conditional generative adversarial network. A conditional generative adversarial network consisting of four parts is obtained by training a large number of images of different age groups marked with ages, and the conditional generative adversarial network comprises an image generator G, an image discriminator D, an age estimation network AEN and an identity recognition network FRN. Wherein G is trained to generate an aged image, and the aged image is automatically and effectively generated by inputting the young image and a preset age condition. And D, identifying whether the generated old image is a real image or not, and ensuring that the generated old image has deception. Wherein the AEN is used for reducing the difference between the age of the generated image and a preset value, and the FRN is used for ensuring the consistency of portrait identities in the generation process. Through the design of the network structure, end-to-end training of the whole network is achieved, face aging is well shown, and high-quality face aging images with the advantages of identity consistency, high cheating performance, high resolution and the like can be generated.

The invention discloses a face age estimation method performing measurement learning based on a convolutional neural network. The method comprises the overall steps that a dataset is constructed; thedataset is divided into a training set and a verification set; paired construction is performed on mini-batches on a network input layer, and then the mini-batches are sent into two twin networks fortraining; a VGG-16 network is constructed; network training is performed; softmax loss and revised contrastive loss are jointly used as supervisory signals to perform network adjustment; network evaluation is performed; and the finally estimated age is a maximum probability corresponding category obtained on a softmax layer. According to the method, deep learning and measurement learning are combined; by introducing measurement learning, the distinction degree of a feature space is higher, and therefore the robustness of an age estimation algorithm is higher; and deep learning is utilized to combine a feature extraction task and an objective function optimization task, end-to-end training is realized for the whole task, and good performance can be obtained when the method is applied to a public dataset.

The invention discloses an age estimation method based on a multi-output convolution neural network and ordered regression. The method comprises the following steps of 1, establishing an Asian face age data set (AFAD); 2, establishing training data used for a dichotomy; 3, training a depth convolution neural network; 4, inputting a test sample into a trained convolution neural network; and 5, acquiring age estimation of the test sample. The invention provides a method of sorting the ages. The ordered regression and a deep learning method are combined so that accuracy of age prediction performance is greatly increased. In an existing age estimation method, characteristic extraction and regression modeling are performed independently and optimization is insufficient. By using the method of the invention, the above problems are solved; a sequence relation of age labels can be fully used to carry out ordered regression of age estimation; age estimation accuracy is increased; a large scale database is established for the age estimation of the Asian faces and a database basis is provided for face age estimation research. The method can be widely used for age estimation of face images.

The invention discloses a three-dimensional human face change simulation method which comprises the steps of constructing a three-dimensional facies cranii database; standardizing a facies cranii model; extracting a human face aging and weight change rule; and simulating the human face aging and weight change. According to the three-dimensional human face change simulation method, aging and weight change are simultaneously considered; and human experience change caused by age increase and weight change can be simulated. According to the three-dimensional human face change simulation method, simulation for aging and weight change of the three-dimensional human face is firstly realized through utilizing facies cranii CT data. According to the three-dimensional human face change simulation method, on condition that three-dimensional time sequence human face data are in shortage, facies cranii data of different persons can be used; the effects of different persons are eliminated from the facies cranii data; and furthermore the aging role and the weight change role of the human face can be extracted. Furthermore an algorithm which is adopted in the three-dimensional human face change simulation method has advantages of simple and high-efficiency operation, and accurate simulation result. The three-dimensional human face change simulation method can be used for searching criminals who have absconded for many years in criminal investigation. The three-dimensional human face change simulation method can be performed in association with medical cosmetic surgery. Furthermore the three-dimensional human face change simulation method can assist cosmetic design, etc. in film and television entertainment.

The invention discloses a face age estimation method capable of carrying out distributed learning on the basis of a convolutional neural network. The overall process of the invention includes the following steps that: data are extracted so as to form an age data set; the age data set is divided into a training set and a verification set; the last fully-connected layer of a deep neural network is followed by a softmax layer; age estimation network training is performed; softmax loss and mean-variance loss are used together as supervised signals so as to adjust the network; the trained network model is evaluated, so that a network model with optimal performance can be selected; and age prediction is carried out on the basis of the obtained model. According to the method of the invention, thenew supervised signals, namely the mean-variance loss, are adopted, and therefore, the nature of correlation between ages is fully utilized, operation such as the manual introduction of variance is avoided, and any manual intervention except for preprocessing is not required.

The invention discloses a human face image age changing method based on average face and aging scale map, comprising the steps of: dividing the human face to be shape vector and vein vector with dense feature presentation; using a human face cartoon synthetic technique to complete young changing of specific human face by calculating difference between the specific human face and the average human face and further reducing the difference; and combining human face image vector modeling with human face image ageing changing for realizing human face image ageing analog in accordance with physiological feature. According to the changing of human face shape feature caused by aging, some feature line of the human face shape is modified so as to realize shape ageing analog; and for the change of vein feature, an aging scale map is prepared in a way of scale map, according to vein mapping, the vein feature aging analog of the human face shape is realized. The method fully considers the unified change of the shape and the vein when human face age changes, thereby greatly increasing the third dimension of synthetic image.

A computer implemented method and apparatus for mitigating face aging errors when performing facial recognition. The method comprises receiving an indication of a face that needs to be searched in an image set, where each image in the image set comprises a timestamp that identifies a creation date of the image, the creation date being in a continuum of successive time intervals; and identifying the indicated face in images taken in each time interval of a plurality of successive time intervals for the indicated face, wherein each face found in images taken in a previous successive time interval is used as a reference set for identifying the face in images taken in a next successive time interval.

The invention provides a single-sample human face recognition method compatible for human face aging recognition, which comprises the steps of conducting the aging simulation on the pre-stored image model of a human face sample to re-construct the image model of the human face sample; conducting the global feature matching for a to-be-recognized human face image model with the image model of the human face sample, wherein if the matching fails, regarding the recognition result as mismatching; and conducting the local feature matching for the to-be-recognized human face image model with the image model of the human face sample, wherein if the matching fails, regarding the recognition result as mismatching. The above to-be-recognized human face image model is an active appearance model of a to-be-recognized human face image. The image model of the human face sample is an active appearance model of a reserved human face sample image. According to the technical scheme of the invention, the recognition effect compatible for human face aging influence is realized and improved based on the combination of the AAM technique with the IBSDT technique. Meanwhile, based on the combination of the AAM technique with the triangulation matching technique, the matching reliability of global features is greatly improved. Based on the combination of the LBP technique with the SURF technique, the matching reliability of local features and the illumination robustness are improved. Finally, the high recognition rate for the reserved human face image as a single sample is realized.

The invention discloses a three-dimensional face age classifying device and method based on three-dimensional point cloud. The device comprises a characteristic region detection unit for positioning a characteristic region of the three-dimensional point cloud, a mapping unit for mapping the three-dimensional point cloud into depth image space, a characteristic calculating unit for calculating the depth image presentation characteristics of a mapped depth image and an age classifier calculating unit for carrying out age classifying based on the depth image presentation characteristic. The method includes the steps of detecting the characteristic region, mapping the depth image, calculating the image presentation characteristic, and carrying out classifying. According to the device and method, the mode that an image presentation characteristic pool is built in cooperation with a plurality of textural characteristics is used, the characteristics of a three-dimensional depth face image are accurately described, then accurate classifying is achieved through an age random forest classifier on the basis of an image presentation characteristic set, and the classifying accuracy is high.

The present invention provides a face age estimation method based on deep learning. The method comprises: S1: establishing a deep learning network model; S2: employing a classification mode to perform pre-training of the deep learning network model to allow the deep learning network model to have a classification capacity; S3: performing fine regulation on the basis of the step S2, and allowing the deep learning network model to have capabilities of learning of an appearance age and estimation of the appearance age; S4: using 80% of a data set in a real age data set as a training set to perform fine regulation of the deep learning network model, and using the 20% of the data set in the real age data set as a test set to perform the test of the deep learning network model; S5: constructing an indirect regression method to estimate an age value on the basis of the software output value of the last one full connection layer in a convolutional neural network model; and S6: inputting a detected face image, and obtaining the face age of the detected face image.

The invention discloses a method for estimating human face age based on GAN extended multi-race feature cooperative selection. Firstly, simulating generation of multi-style human face samples is carried out through a generating antagonism network, so as to rapidly and large-scale expand human face databases of different races, thereby improving the recognition accuracy of age information of yellow, brown and other races. Then the convolution neural network is used to pre-train the original dataset, and then further refined training is carried out based on the expanded face age database. And finally four people of Sub-CNN performs joint feature selection fusion based on group sparse algorithm to solve the problem of age estimation based on face image. The invention obtains a face age estimation model with more generalization ability, and at the same time, the invention can greatly improve the performance of a face recognition system of many ages, and makes up for the shortcomings of theprevious research.

The invention provides an Asian human face age characteristic model generating method, and the method comprises the steps: S1), extracting a wrinkle difference characteristic pattern of each human face image in a training set, and obtaining an original feature vector of each human face image; S2), determining the number of reduced dimensions of each original feature vector, reducing the dimensions of the original feature vector of each original image based on a PCA method, and obtaining feature vectors after dimension reduction; S3), carrying out the training of the feature vectors based on a support vector machine regression algorithm after dimension reduction, and generating an age characteristic model. In addition, the invention also provides the age characteristic model generated through the above method, and an Asian human face age estimation method. The age characteristic model carries out the extraction of the features based on a human face important organ region and a human face wrinkle region, and guarantees that the features comprise the sufficient information sensitive to the age estimation. The age estimation method can obtain a more precise age estimation value, and meets the demands of an actual application scene.

The invention provides a human face age estimation method based on a deep sparse representation, and belongs to the technical field of image processing and pattern recognition. The method solves the problem that an existing human face age estimation method is unstable. The method mainly comprises the following steps: A, building a distinguishing dictionary learning model; B, establishing a deep sparse representation model based on a distinguishing dictionary; C, building a two-factor analysis model, and removing an identity factor; D, extracting a robustness age characteristic; and E, building a stratified age estimation model, and performing age estimation. The method has the advantages of high anti-interference capability, high accuracy and the like.

The invention discloses a face age bracket recognition method, and the method comprises the steps: (a), obtaining the face features of all face images in a training sample set; (b), carrying out the pre-training of a multilayer stack-type self-coding model; (c), carrying out the coding of the face features of all face images, and obtaining the age bracket features of all face images; (d), carryingout the clustering of the age bracket features of all face images, and obtaining a preset number of clustering centers; (e), calculating the membership degree of the age bracket features of all faceimages to all clustering degrees, and adjusting the network parameters of the multilayer stack-type self-coding model, so as to optimize the membership degree; (f), repeatedly carrying out the steps (c)-(e) till a preset condition is met; (g), carrying out the coding of a to-be-processed face image, and obtaining the age bracket features of the to-be-processed face image; (h), carrying out the agebracket recognition of the to-be-processed face image, and obtaining the age bracket type of the to-be-processed face image. The invention also provides a face age bracket recognition device, a computer device, and a readable storage medium. The method can achieve the quick and efficient recognition of a face age bracket.

The present invention discloses a dynamic interval-based face age estimation method. The method comprises the steps of extracting the features of a face; finding a central face at each age based on the clustering algorithm; according to the similarity degree between a to-be-estimated face and the central face, selecting top K ages corresponding to a most similar face as prediction ages and defining a correlation function between the confidence interval and the confidence coefficient; according to the correlation function between the confidence interval and the confidence coefficient, estimating the bar graph, the normal distribution curve, the normal distribution expected value and the standard deviation of a face age according to the square estimation method by utilizing the top K ages of all test images at each age; calculating different confidence coefficients and the corresponding confidence intervals thereof according to the cumulative distribution function of the normal distribution. According to the technical scheme of the invention, the problem in the prior art that the traditional single-age estimation method is not high in accuracy can be effectively solved.

The invention discloses a method and a device for simulating human face aging based on homologous continuity. The shape feature vector is constructed according to the feature points and the metric iscalculated. The metric and the preset metric are matched and calculated to obtain the corresponding matching face images of each age group which are equal to the preset number. A texture enhancement face prototype is acquired corresponding to each age group; the source target age range of the face image is determined; according to the shape feature vectors, age-related and target shape feature vectors corresponding to the face prototype, the first simulated image simulating the aging of the face shape is obtained. According to the texture feature vector constructed by the pixel points corresponding to the face prototype, the second simulated image which simulates the aging of face texture is obtained. The device comprises: an electronic device which is communicated by a processor, a memoryand a bus; and a non-transient computer-readable storage media. The device performs the method described above. The method and the device can truly and naturally simulate the aging of a human face.

The invention discloses a face age estimation method and a device based on deep learning. The device comprises a face recognition module, a preprocessing module and a human face age estimation module.The face recognition module is used for recognizing a face region image from an input image. The preprocessing module is used for preprocessing the face region image, and adjusting the face area image to be a face area image of a front face posture. The human face age estimation module is used for inputting the preprocessed face region image into a convolution neural network model and outputtinga face age estimation result. According to the method and the device, the convolution neural network model is established and the human face age estimation can be achieved. The estimation result is more accurate.

The invention discloses a face aging image synthesis method based on a cyclic conditional generative adversarial network, and belongs to the field of computer vision. The method comprises the steps offirstly selecting a generative adversarial network as a basic framework; meanwhile, taking the idea of dual learning of the cyclic generative adversarial network as reference; utilizing a supervisedlearning thought of an auxiliary discriminator; introducing a category label innovatively when the cyclic generative adversarial network is used for generating an aging picture; according to the method, the network is enabled to increase attention to specific age characteristics, an auxiliary classification branch is added to the discriminator so that the generation network is enabled to effectively utilize label information to learn specific knowledge, and generation and conversion of the generation network in images of different age groups can be completed through single training via the idea of dual learning. Through the method, the advantages of dual learning and auxiliary classification supervision thought are fully utilized, and the efficiency and picture quality of the cyclic generative adversarial network in senescence image generation are greatly improved.

The invention, which belongs to the technical field of face image comparison in criminal investigation, especially relates to an improved face image comparison method. The method is characterized in that the method comprises: S1, holographic shooting is carried out; S2, depth data filtering is carried out on the holographic picture; S3, face detection is carried out by using a Viola-Jones method and face segmentation is carried out by using a threshold method; S4, face tracking and point cloud registration are carried out to obtain a complete 3D face image model; S5, multi-attitude multi-illumination image synthesis is carried out to obtain a front attitude face image; S6, key feature extraction is carried out; S7, feature comparison is carried out to realize person identification; and S8, face age deduction is carried out. Therefore, the restriction that the face inclination degree of the image employed by the machine face comparison system must be less than 15 degrees can be broken; and omnibearing face image comparison can be carried out at multiple angles. Compared with the prior art, the provided method enables the comparison condition and requirement to be relaxed substantially and the comparison accuracy to be improved substantially.

The present invention discloses an adaptive age distribution learning-based face age estimation method. The method comprises the following steps that: step 1, required face image data are provided; step 2, the age distribution of face samples is established; step 3, a face age prediction model is established; step 4, the optimized objective function of an algorithm is established; and step 5, the prediction model is utilized to estimate the age tag of a face image. With the method adopted, the internal structure of the face sample can be effectively grasped, and age tag ambiguity is analyzed through using the age tags of the context-dependent samples.

The invention discloses a face image processing model training method and device, electronic equipment and a storage medium, and the method comprises the steps: carrying out the coding of a first faceimage through an encoder of a face image processing model, and obtaining a first feature vector of the first face image; splitting the first feature vector into a plurality of sub-feature vectors, wherein the sub-feature vectors at least comprise an identity sub-feature vector and an age sub-feature vector; determining a to-be-decoded vector according to each sub-feature vector, and decoding theto-be-decoded vector by using a decoder of the face image processing model to obtain a second face image; and optimizing parameters of the face image processing model according to the regression lossvalues of the first face image and the second face image. According to the face image processing model obtained through training, the dependence on data distribution is reduced, the method is more robust to long-tail data with unbalanced ages, and a face aging image with a better effect can be generated.