相关论文
2019 - Canadian Conference on AI
Fully End-To-End Super-Resolved Bone Age Estimation

With the release of large-scale bone age assessment datasets and competitions looking at solving the problem of bone age estimation, there has been a large boom of machine learning in medical imaging which has attempted to solve this problem. Although many of these approaches use convolutional neural networks, they often include some specialized form of preprocessing which is often lengthy. We propose using a subpixel convolution layer in addition to an attention mechanism similar to those developed by Luong et al. in order to overcome some of the implicit problems with assuming particular placement and orientation of radiographs due to forced preprocessing.

2019
Age Estimation and Gender Prediction Using Convolutional Neural Network

A face of a person plays a very crucial part to identify the person uniquely. Two important traits namely age and gender can enhance the performance of face recognition. From the human perspective, it is very easy to estimate the person’s gender and age on the basis of visualization but if we see from the machine perspective they can’t. To estimate the age and gender from the facial image is a challenging task for a machine due to variations, lighting and other conditions in the face. This paper proposed the estimation of age and predict the gender of a person from the facial image using the convolutional neural network. In the proposed methodology before the training and testing, also apply the PCA to reduce the extracted features dimensions. This work is done on the publicly available IMDB-WIKI dataset as well as own dataset using the MATLAB platform for the implementation purpose.

2016 - Proceedings of the 24th ACM international conference on Multimedia
Facial Age Estimation Using Robust Label Distribution

Facial age estimation, to predict the persons' exact ages given facial images, usually encounters the data sparsity problem due to the difficulties in data annotation. To mitigate the suffering from sparse data, a recent label distribution learning (LDL) algorithm attempts to embed label correlation into a classification based framework. However, the conventional label distribution learning framework only considers correlations across the neighbouring variables (ages), which omits the intrinsic complexity of age classes during different ageing periods (age groups). In the light of this, we introduce a novel concept of robust label distribution for scalar-valued labels, which is designed to encode the age scalars into label distribution matrices, i.e. two-dimensional Gaussian distributions along age classes and age groups respectively. Overcoming the limitations of conventional hard group boundaries in age grouping and capturing intrinsic inter-group dependency, our framework achieves robust and competitive performance over the conventional algorithms on two popular benchmarks for human age estimation.

1997
Estimation of Age for Trachemys scripta and Deirochelys reticularia by Counting Annual Growth Layers in Claws

The need for accurate methods of age determination in life-history studies is well recognized. Techniques used to estimate the age of turtles include analysis of body size (Balazs, 1982; Frazer and Ehrhart, 1985; Frazer and Ladner, 1986), weighing eye lenses (. Frazier,J. Ballon, and S. Salas, unpubl.), counting growth zones on plastral scutes (Cagle, 1946; Sexton, 1959; Gibbons, 1983), and skeletochronology (Mattox, 1935; Zug, 1990; Klinger and Musick, 1992). Each technique has its disadvantages, and finding a technique that is accurate and can be applied objectively has been difficult (Zug, 1991). An ideal age estimation technique would not cause stress or mortality to the individual being aged, would be valid for all age/size classes, would be based on a clearly recognizable structure that is permanent throughout the life of the individual, and would not require subjective judgment. The use of claws could potentially satisfy three of these requirements for turtle age estimation. Zug (1991) suggested that because claws grow continuously throughout the life of a turtle and because growth is affected by environmental factors, the formation of seasonal growth layers might be expected. The objective of our study was to determine whether annual layers could be observed in sections of claws from Trachemys scripta and Deirochelys reticularia and whether counts based on such layering could provide reliable data for age estimation.

2019 - Finanzarchiv
Inferring Age at Death for Japanese and Thai Skeletal Samples under a Bayesian Framework of Analysis: A Test of Priors and Their Effects on Estimation

A critical step in age-at-death estimation is to identify the age-at-death distribution to which the unknown skeletal remains most likely belong. Age estimation based on a frequentist approach assumes that the age distribution of the target population is same as that of the reference sample. In a Bayesian framework, researchers have greater flexibility, with the freedom to specify mortality information via a prior distribution and integrate it with osteological data to produce more accurate age estimates. The selection of an optimal prior remains challenging, as forensic anthropologists analyze unidentified individuals often originating from unknown populations. Understanding the effects of the prior choice on the final age produced for various populations is essential to the interpretation of these estimates. In this article we investigate the effects of three different priors on age estimates using scores of three traditional methods collected on Asian skeletal samples representing peoples from Japan and Thailand. We test a uniform prior, which assumes the equal chance of death regardless of age, as well as various informative priors, derived from a Japanese mortality database and skeletal collections. We combine each of the priors with the parameters of the cumulative probit regression model to obtain age estimates. The results of our analyses show that the informative prior outperforms when it is carefully chosen to reflect the geographic and temporal origins of the target population. While the uniform prior produces the least-biased age estimates, one should be cautious, as it can generate unrealistically old and inaccurate ages.

2017 - 2017 2nd International Conference on Bio-engineering for Smart Technologies (BioSMART)
Human age estimation using auditory system through dynamic frequency sound

Age estimation has recently become increasingly important for a variety of reasons including security, gaming, healthcare, and biometry. In this paper, we are interested in age estimation for content filters which could limit access to age-inappropriate content. Considering that the auditory response to sound varies with age, we provide in this paper preliminary insights about age estimation through the response of the human auditory system to dynamic frequency sound. The net results are very encouraging and promising.

2013 - Multimedia Tools and Applications
Effective human age estimation using a two-stage approach based on Lie Algebrized Gaussians feature

Automatically and effectively estimating human ages via facial images has lots of practical applications, such as security surveillance, electronic customer relationship management and entertainment. Motivated by the fact that feature representation and recognition are two key problems in facial image based human age estimation, in this paper, we propose to employ a novel discriminative feature called Lie Algebrized Gaussians (LAG) for the representation of age images and design a two-stage approach for learning and predicting human ages. LAG is built on Gaussian Mixture Models (GMM) and is able to capture the aging manifold of the age image by preserving the Lie group manifold structure information embedded in the feature space. Given the LAG feature for each image, we estimate the human age using a two-stage approach in a coarse-to-fine fashion. In the first stage, an adaptive age group for each input image is obtained by selecting a number of neighboring age labels around the output of a global regressor. In the second stage, a local classifier is learned from the selected age classes to determine the final age of the input image. The effectiveness of our approach is evaluated on both FG-NET and MORPH benchmarks, extensive experimental results and comparisons with the state-of-the-art algorithms demonstrate the superiority of our approach for the human age estimation task.

2015 - Psychological Reports
Examining the Effect of Smile Intensity on Age Perceptions

Research has demonstrated the positive effects of smiles on interpersonal perceptions of attractiveness, likability, and friendliness. A possible mechanism underlying the effects of smiles is babyfacedness. Four studies were conducted with 1,235 participants. In Study 1, 646 participants were assigned to one of the six levels of smile intensity and responded to the measures of age perception and perceived babyfacedness. Compared to the neutral expression, the maximal smile reduced age estimations and this effect was mediated by perceived babyfacedness. In Study 2, 59 respondents' responses indicated that a maximal smile reduced the perception of age. In Study 3, 318 respondents estimated the age of models in different sex and levels of smile intensity. Maximal smiles reduced age estimations. In Study 4, 125 students and 87 non-students were randomly assigned to one of the three levels of smile intensity and provided age estimations. Replicating previous findings, maximal smiles reduced age estimations. Consistent results emerged from the various contexts, models, and samples, as well as in a pooled analysis.

2016 - 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI)
From individual hand bone age estimates to fully automated age estimation via learning-based information fusion

Increasingly important for both clinical and forensic medicine, radiological age estimation is performed by fusing independent bone age estimates from hand images. In this work, we show that the artificial separation into bone independent age estimates as used in established fusion techniques can be overcome. Thus, we treat aging as a global developmental process, by implicitly fusing developmental information from different bones in a dedicated regression algorithm. With 0.82 ± 0.56 years absolute deviation from chronological age on a database of 132 3D MR hand images, the results of this novel automatic algorithm are inline with radiologists performing visual examinations.

2010 - 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops
A robust framework for multiview age estimation

Age estimation from facial images has promising applications in human-computer interaction, biometrics, visual surveillance, and electronic customer relationship management, etc. Most existing techniques and systems can only handle frontal or near frontal view age estimation due to the difficulties of 1) differentiating diverse variations from uncontrollable and personalized aging patterns on faces and 2) collecting a fairly large database covering the chronometrical image series for each individual in different views. In this paper, we propose a robust framework to deal with multiview age estimation problem. A large face database, with significant age, pose, gender, and identity variations, is exploited in the experiments. In our framework, the training set is partitioned into small groups, namely code groups, according to their multiple labels, e.g. pose and age. Based on certain set-set distance measure, a compact representation for each image is obtained by measuring the distance between the image and all the code groups, which can be followed by classification or regression algorithms. Extensive experiments and comparisons with traditional multiview models demonstrate the proposed framework with significant advantages of variation decomposable, classifier adaptable, and feature selectable and extendable.

2019 - Radiology. Artificial intelligence
Improving Automated Pediatric Bone Age Estimation Using Ensembles of Models from the 2017 RSNA Machine Learning Challenge.

Purpose To investigate improvements in performance for automatic bone age estimation that can be gained through model ensembling. Materials and Methods A total of 48 submissions from the 2017 RSNA Pediatric Bone Age Machine Learning Challenge were used. Participants were provided with 12 611 pediatric hand radiographs with bone ages determined by a pediatric radiologist to develop models for bone age determination. The final results were determined using a test set of 200 radiographs labeled with the weighted average of six ratings. The mean pairwise model correlation and performance of all possible model combinations for ensembles of up to 10 models using the mean absolute deviation (MAD) were evaluated. A bootstrap analysis using the 200 test radiographs was conducted to estimate the true generalization MAD. Results The estimated generalization MAD of a single model was 4.55 months. The best-performing ensemble consisted of four models with an MAD of 3.79 months. The mean pairwise correlation of models within this ensemble was 0.47. In comparison, the lowest achievable MAD by combining the highest-ranking models based on individual scores was 3.93 months using eight models with a mean pairwise model correlation of 0.67. Conclusion Combining less-correlated, high-performing models resulted in better performance than naively combining the top-performing models. Machine learning competitions within radiology should be encouraged to spur development of heterogeneous models whose predictions can be combined to achieve optimal performance.© RSNA, 2019 Supplemental material is available for this article. See also the commentary by Siegel in this issue.

2017 - International Journal of Legal Medicine
New approach to age estimation of male and female adult skeletons based on the morphological characteristics of the acetabulum

Age estimation is essential to the human identification process, both in forensic and archeological contexts. Based on the previous male-specific method of Rissech et al. (J Forensic Sci, 2006, 51:213–229), a new approach to age estimation based on the acetabulum has been described and evaluated in 611 individuals from the Lisbon Collection (Lisbon, Portugal). This paper has two main goals: (1) to revise and better define the variables of Rissech’s method related to the acetabular fossa, namely, variables 5, 6, and 7, and (2) to extend the applicability of this new approach to both sexes while analyzing age-related sex differences in the acetabular aging process. The results demonstrate the suitability of combining acetabulum traits and a Bayesian approach to estimating age in adults of both sexes. This study has confirmed the usefulness of the redefined variables of the acetabular fossa when age-related changes are considered. Furthermore, the newly defined variables have good to excellent values of repeatability. The study has also extended the method’s applicability to females. The revised method has absolute error averages of 7.28 years for males and 7.09 years for females, based on a sex-specific reference sample. In addition, approximately 74 % of the individuals estimated had an absolute error less than 10 years. Interestingly, the acetabular aging process follows similar trends in both sexes, but the aging rate seems to be different between males and females, especially in middle-aged individuals. Despite the fact that the age estimates, on average, did not improve significantly with the use of a sex-specific reference sample, it is recommended that the sexes be analyzed separately due to the differences in aging rates and inaccuracy values.

1989
The Estimation of Tooth Age from Attrition of the Occlusal Surface

Age estimation in unidentified bodies is inaccurate. Usually only a broad range of ages, such as 20–30 years or 30–50 years, can be given, especially when postmortem change has occurred. Thus there is a real need in routine forensic practice for greater accuracy. Takei (1970, 1981), looked at the relationship between teeth and age by using the ‘Theory of Quantification’. Song and Jia (1987). extended the use of this technique, adding multiple stepwise regression analysis, to the statistical examination of data derived from the study of attrition of the occlusal surfaces of teeth. The computer model thus derived provides a basis for an accurate assessment of age from qualitative material. This paper discusses the technique and is based upon the findings made during a study of the degree of occlusal attrition in 880 people, living in rural and urban areas of the Liaoning province of the People's Republic of China.

2003
Postnatal Development in the Indian Short-Nosed Fruit Bat Cynopterus sphinx: Growth Rate and Age Estimation

We studied the patterns of postnatal growth and changes in length of forearm, body mass and total epiphyseal gap in the captive free-flying short-nosed fruit bat Cynopterus sphinx. At birth young were altricial. By day five, their eyes had opened, and the pinnae become unfolded between sixth and ninth day of age. At the age of three days, the mean forearm length and body mass were equivalent to 42.2% and 18.2%, respectively of the values of postpartum females. The length of forearm increased linearly until 36 days and attained 94.1% of mean forearm length of postpartum females at the age of 219 days. Body mass of pups increased linearly until 60 days and attained 72.7% of mean mass of postpartum females at the age of 219 days. The length of total epiphyseal gap of fourth metacarpal-phalangeal joint increased until 15 days of age and subsequently decreased linearly, and closed at about 60 days of age. The age predicting equation based on the length of forearm is valid when its dimensions are between 29.4 mm and 52.4 mm (3–36 days of age). Similar equation but based on the length of total epiphyseal gap is valid when its dimensions range from 47.0 µm to 6.0 µm (15–60 days of age). Growth patterns of forearm length and body mass were best described by the logistic and Gompertz nonlinear growth models, respectively. There was no significant difference in the growth patterns of body mass and length of total epiphyseal gap with reference to lengths of forearm of captive and wild-grown pups.

2013 - Journal of forensic and legal medicine
The French-Canadian data set of Demirjian for dental age estimation: a systematic review and meta-analysis.

BACKGROUND Estimation of age of an individual can be performed by evaluating the pattern of dental development. A dataset for age estimation based on the dental maturity of a French-Canadian population was published over 35 years ago and has become the most widely accepted dataset. The applicability of this dataset has been tested on different population groups. AIM To estimate the observed differences between Chronological age (CA) and Dental age (DA) when the French Canadian dataset was used to estimate the age of different population groups. MATERIALS AND METHODS A systematic search of literature for papers utilizing the French Canadian dataset for age estimation was performed. All language articles from PubMed, Embase and Cochrane databases were electronically searched for terms 'Demirjian' and 'Dental age' published between January 1973 and December 2011. A hand search of articles was also conducted. RESULTS A total of 274 studies were identified from which 34 studies were included for qualitative analysis and 12 studies were included for quantitative assessment and meta-analysis. When synthesizing the estimation results from different population groups, on average, the Demirjian dataset overestimated the age of females by 0.65 years (-0.10 years to +2.82 years) and males by 0.60 years (-0.23 years to +3.04 years). CONCLUSION The French Canadian dataset overestimates the age of the subjects by more than six months and hence this dataset should be used only with considerable caution when estimating age of group of subjects of any global population.

2005 - International Journal of Legal Medicine
The influence of slice thickness on assessment of clavicle ossification in forensic age diagnostics

Computed tomography scans originally obtained to assess the ossification status of the medial clavicular epiphyses of 40 live subjects for forensic age estimation purposes were analysed. From the data acquired we reconstructed images with slice thicknesses of 1, 3, 5 and 7 mm, and based on the classification of stages by Schmeling et al. (Int J Legal Med 118:5–8, 2004), we determined the ossification stage for each reconstructed slice thickness, separately for both sides. Of the 80 clavicular epiphysial plates examined, seven displayed slice-thickness-dependent differences at certain ossification stages. In one case a slice thickness of 1 mm led to a different diagnosis of the ossification stage than a slice thickness of 3 mm, in three cases the diagnoses differed between the slice thicknesses of 3 and of 5 mm, and in another three cases, between 5 and 7 mm. We therefore conclude that for age estimation purposes, the slice thickness should be 1 mm to ensure maximum accuracy and diagnostic reliability.

2016 - Journal of forensic and legal medicine
Dental age assessment in a northern Chinese population.

INTRODUCTION Age estimation is imperative in the fields of paediatric dentistry, orthodontics and forensic science. Studies have shown that dental age estimation by the radiological method is reliable and non-destructive. Although Demirjian's method was the most widely used estimation method, in recent studies, the Willems' method has been found to be more accurate. The aim of this study was to compare the accuracy of dental age estimation methods and to modify the Demirjian method to make it more applicable for a northern Chinese population. MATERIALS AND METHODS An assessment was made of 1004 digital orthopantomographs of a northern Chinese population (392 boys and 612 girls) ranging in age from 11 to 18 years old. Dental ages were calculated using both the Demirjian method and the Willems method. Discrepancies between chronological ages and dental ages were statistically analysed by the paired t-test and the Wilcoxon signed rank test. A nonlinear fitting method was applied to construct a mathematical model to modify the Demirjian method. RESULTS The Demirjian method underestimated age by 0.47 y in boys and 0.63 y in girls, while the Willems method underestimated age by 0.54 y and 1.01 y in boys and girls, respectively. The mean absolute error was 1.08 y for the Demirjian method and 1.22 y for the Willems method. CONCLUSION The Demirjian method was more accurate for estimating dental age compared with the Willems method. However, the Demirjian method may not be suitable for the northern Chinese population; therefore, it should be modified so that it can be used for this population.

2009 - CVPR
Human age estimation using bio-inspired features

We investigate the biologically inspired features (BIF) for human age estimation from faces. As in previous bio-inspired models, a pyramid of Gabor filters are used at all positions of the input image for the S1 units. But unlike previous models, we find that the pre-learned prototypes for the S2 layer and then progressing to C2 cannot work well for age estimation. We also propose to use Gabor filters with smaller sizes and suggest to determine the number of bands and orientations in a problem-specific manner, rather than using a predefined number. More importantly, we propose a new operator “STD” to encode the aging subtlety on faces. Evaluated on the large database YGA with 8,000 face images and the public available FG-NET database, our approach achieves significant improvements in age estimation accuracy over the state-of-the-art methods. By applying our system to some Internet face images, we show the robustness of our method and the potential of cross-race age estimation, which has not been explored by any studies before.

2018 - IEEE Transactions on Information Forensics and Security
An Ensemble CNN2ELM for Age Estimation

Age estimation is a challenging task, because it can be easily affected by gender, race, and other intrinsic and extrinsic attributes. At the same time, performing age estimation for a narrow age range may lead to better results. In this paper, to achieve robust age estimation, an ensemble structure referred to as CNN2ELM, which includes convolutional neural network (CNN) and extreme learning machine (ELM), is proposed for age estimation. The three-level system includes feature extraction and fusion, age grouping via an ELM classifier, and age estimation via an ELM regressor. Age-Net, Gender-Net, and Race-Net are trained using different targets, such as age class, gender class, and race class, respectively, and the three networks are used to extract features corresponding to age, gender, and race from the same image of a person during validation and test stages. Features related to the age property are enhanced by fusing these of race and gender properties. Then, to achieve a narrow age range, the ELM classifies the fusion results into one of the age groups. Afterward, an age decision is made using an ELM regressor. Our network is pretrained on an ImageNet database and then fine-tuned on the IMDB–WIKI database. The recently released Adience benchmark, ChaLearn Looking at People 2016 (LAP-2016), and MORPH-II are used to verify the performance of “Race-Net + Age-Net + Gender-Net + ELM classifier + ELM regressor (RAGN).” RAGN outperforms the existing state-of-the-art age estimation methods. The mean absolute error of the age estimation of RAGN for MORPH-II is determined to be 2.61 years; the accuracy of the age estimation for the Adience benchmark is 0.6649; and the normal score ( $\epsilon$ ) for the sequestered test set of the LAP-2016 data set is 0.3679.

2004 - International Journal of Legal Medicine
Quantification of secondary dentine formation from orthopantomograms—a contribution to forensic age estimation methods in adults

The accuracy of most age determination methods in adults is unsatisfactory and therefore more precise methods are required. Teeth provide several useful points about an individuals age. With advancing age secondary dentine is deposited along the wall of the dental pulp chamber leading to a reduction in the size of the pulp cavity. These age-related changes can be determined and measured from dental radiographs. The aim of this study was to explore if measurements of the size of the pulp cavity performed on orthopantomograms can be used for individual age estimation. Measurements were made digitally for 6 types of teeth from orthopantomograms of 168 individuals aged between 14 and 81 years old. To compensate for errors due to the technique of radiography—for instance differences in angulation—ratios of distances were calculated: pulp/root length, pulp/tooth length and pulp/root width at three different root levels. The width ratios of the pulp cavity showed significant correlation to the chronological age and the coefficient of determination (r2) was highest in the upper lateral incisors (r2=0.913) when an exponential or a logistic regression model was constructed. At the same distance with a linear regression model the coefficient of determination (r2) reached 0.839.

Multi-task warped Gaussian process for personalized age estimation

Dit-Yan Yeung
|
Yu Zhang
|
D. Yeung
|
Yu Zhang

Abstract:Automatic age estimation from facial images has aroused research interests in recent years due to its promising potential for some computer vision applications. Among the methods proposed to date, personalized age estimation methods generally outperform global age estimation methods by learning a separate age estimator for each person in the training data set. However, since typical age databases only contain very limited training data for each person, training a separate age estimator using only training data for that person runs a high risk of overfitting the data and hence the prediction performance is limited. In this paper, we propose a novel approach to age estimation by formulating the problem as a multi-task learning problem. Based on a variant of the Gaussian process (GP) called warped Gaussian process (WGP), we propose a multi-task extension called multi-task warped Gaussian process (MTWGP). Age estimation is formulated as a multi-task regression problem in which each learning task refers to estimation of the age function for each person. While MTWGP models common features shared by different tasks (persons), it also allows task-specific (person-specific) features to be learned automatically. Moreover, unlike previous age estimation methods which need to specify the form of the regression functions or determine many parameters in the functions using inefficient methods such as cross validation, the form of the regression functions in MTWGP is implicitly defined by the kernel function and all its model parameters can be learned from data automatically. We have conducted experiments on two publicly available age databases, FG-NET and MORPH. The experimental results are very promising in showing that MTWGP compares favorably with state-of-the-art age estimation methods.

参考文献

[1]  Sebastian Thrun,et al.  Is Learning The n-th Thing Any Easier Than Learning The First? , 1995, NIPS.

[2]  Rich Caruana,et al.  Multitask Learning , 1998, Encyclopedia of Machine Learning and Data Mining.

[3]  Niels da Vitoria Lobo,et al.  Age Classification from Facial Images , 1999, Comput. Vis. Image Underst..

[4]  Timothy F. Cootes,et al.  Active Appearance Models , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Timothy F. Cootes,et al.  Toward Automatic Simulation of Aging Effects on Face Images , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Neil D. Lawrence,et al.  Fast Sparse Gaussian Process Methods: The Informative Vector Machine , 2002, NIPS.

[7]  Carl E. Rasmussen,et al.  Warped Gaussian Processes , 2003, NIPS.

[8]  Jonathan Baxter,et al.  A Bayesian/Information Theoretic Model of Learning to Learn via Multiple Task Sampling , 1997, Machine Learning.

[9]  Neil D. Lawrence,et al.  Learning to learn with the informative vector machine , 2004, ICML.

[10]  Rama Chellappa,et al.  Facial similarity across age, disguise, illumination and pose , 2004, 2004 International Conference on Image Processing, 2004. ICIP '04..

[11]  C. Christodoulou,et al.  Comparing different classifiers for automatic age estimation , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[12]  Rama Chellappa,et al.  Face Verification Across Age Progression , 2006, IEEE Transactions on Image Processing.

[13]  Yu Zhang,et al.  Learning from facial aging patterns for automatic age estimation , 2006, MM '06.

[14]  Karl Ricanek,et al.  MORPH: a longitudinal image database of normal adult age-progression , 2006, 7th International Conference on Automatic Face and Gesture Recognition (FGR06).

[15]  Rama Chellappa,et al.  Modeling Age Progression in Young Faces , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[16]  Shuicheng Yan,et al.  Learning Auto-Structured Regressor from Uncertain Nonnegative Labels , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[17]  Massimiliano Pontil,et al.  Convex multi-task feature learning , 2008, Machine Learning.

[18]  Zhi-Hua Zhou,et al.  Automatic Age Estimation Based on Facial Aging Patterns , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[19]  Stefano Soatto,et al.  A Study of Face Recognition as People Age , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[20]  Shuicheng Yan,et al.  Ranking with Uncertain Labels , 2007, 2007 IEEE International Conference on Multimedia and Expo.

[21]  Ming Liu,et al.  Regression from patch-kernel , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[22]  Yun Fu,et al.  Image-Based Human Age Estimation by Manifold Learning and Locally Adjusted Robust Regression , 2008, IEEE Transactions on Image Processing.

[23]  Yun Fu,et al.  Human Age Estimation With Regression on Discriminative Aging Manifold , 2008, IEEE Transactions on Multimedia.

[24]  Zhi-Hua Zhou,et al.  Facial age estimation by nonlinear aging pattern subspace , 2008, ACM Multimedia.

[25]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[26]  Heng Tao Shen,et al.  Principal Component Analysis , 2009, Encyclopedia of Biometrics.

[27]  Wen Gao,et al.  Learning long term face aging patterns from partially dense aging databases , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[28]  Thomas S. Huang,et al.  Human age estimation using bio-inspired features , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[29]  Yun Fu,et al.  A study on automatic age estimation using a large database , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[30]  Hongyuan Zha,et al.  Learning distance metric for regression by semidefinite programming with application to human age estimation , 2009, ACM Multimedia.

[31]  Kate Smith-Miles,et al.  Facial age estimation by multilinear subspace analysis , 2009, 2009 IEEE International Conference on Acoustics, Speech and Signal Processing.

[32]  John K Kruschke,et al.  Bayesian data analysis. , 2010, Wiley interdisciplinary reviews. Cognitive science.

引用
Novel Nonlinear Approach for Real-Time Fatigue EEG Data: An Infinitely Warped Model of Weighted Permutation Entropy
IEEE Transactions on Intelligent Transportation Systems
2020
Computational Intelligence in Automatic Face Age Estimation: A Survey
IEEE Transactions on Emerging Topics in Computational Intelligence
2019
Multi-step prediction with missing smart sensor data using multi-task Gaussian processes
2017 IEEE International Conference on Big Data (Big Data)
2017
Improving Face-Based Age Estimation With Attention-Based Dynamic Patch Fusion
IEEE Transactions on Image Processing
2022
Fuzzy Bin-Based Classification for Detecting Children’s Presence with 3D Depth Cameras
ACM Trans. Sens. Networks
2017
Relative Attribute SVM+ Learning for Age Estimation
IEEE Transactions on Cybernetics
2016
Local Rademacher Complexity-based Learning Guarantees for Multi-Task Learning
J. Mach. Learn. Res.
2016
Anchored Regression Networks Applied to Age Estimation and Super Resolution
2017 IEEE International Conference on Computer Vision (ICCV)
2017
Deeply Learned Rich Coding for Cross-Dataset Facial Age Estimation
2015 IEEE International Conference on Computer Vision Workshop (ICCVW)
2015
Improving Quadrature for Constrained Integrands
AISTATS
2018
Age Estimation via Grouping and Decision Fusion
IEEE Transactions on Information Forensics and Security
2015
Learning Mid-Level Features and Modeling Neuron Selectivity for Image Classification
ArXiv
2014
Modeling Neuron Selectivity Over Simple Midlevel Features for Image Classification
IEEE Transactions on Image Processing
2015
Quantifying Facial Age by Posterior of Age Comparisons
BMVC
2017
Bayesian Warped Gaussian Processes
NIPS
2012
Feature-based transfer learning with real-world applications
2010
Facial Image-based Age Estimation
2010
3D Age Progression Prediction in Children's Faces with a Small Exemplar-Image Set
J. Inf. Sci. Eng.
2014
3-D Age Progression Prediction in Children ’ s Faces with a Small Exemplar-Image Set C
2012
Facial age estimation based on label-sensitive learning and age-oriented regression
Pattern Recognit.
2013