Age and Gender Estimation of Unfiltered Faces

By Eran Eidinger, Roee Enbar, Tal Hassner. 2014

793 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

Designing Disaggregated Evaluations of AI Systems: Choices, Considerations, and Tradeoffs (2021, methodology): “Reusing an existing dataset, such as the IARPA Janus Benchmark A (IJB-A) dataset for evaluating face-detection and face-recognition AI systems [77] or the Adience dataset for evaluating age and gender classifiers [22], is less time consuming and less expensive than creating a new dataset.”
Exposing and Correcting the Gender Bias in Image Captioning Datasets and Models (2019, methodology): “Classifying gender as well as age from images of faces using CNNs is also common [Levi and Hassner, 2015], [Eidinger et al., 2014], [Gonz´alez-Briones et al., 2018], [Dehghan et al., 2017], [Liu et al., 2017].”
Women also Snowboard: Overcoming Bias in Captioning Models (2018, background): “Gender classification models frequently focus on facial features [38,39,40].”
Towards fairer datasets: filtering and balancing the distribution of the people subtree in the ImageNet hierarchy (2019, background): “Examples range from quantifying beauty [47] to predicting sexual orientation [76] to (arguably) annotating gender [25, 42].”
Diversity in Faces (2019, methodology): “Many other data sets have been proposed for different facial analysis tasks, such as age and gender classification [24, 27, 31, 54–56], facial expression analysis [55], memorability [57], attributes [35] and aging [58].”
Compacting, Picking and Growing for Unforgetting Continual Learning (2019, positive): “Finally, Adience dataset [7] contains faces with labels of eight different age groups.”
The MegaFace Benchmark: 1 Million Faces for Recognition at Scale (2015, methodology): “[9] created a dataset of 27K photos of 2.”
Deep Multi-task Representation Learning: A Tensor Factorisation Approach (2016, background): “Dataset, Settings and Baselines The AdienceFaces (Eidinger et al., 2014) is a large-scale face images dataset with the labels of each person’s gender and age group.”
Towards Measuring Fairness in AI: The Casual Conversations Dataset (2021, background): “Levi & Hassner [17] proposed an end-to-end trained Convolutional Neural Network (CNN) on Adience benchmark [8] to predict apparent age and gender from faces.”
AFIF4: Deep Gender Classification based on AdaBoost-based Fusion of Isolated Facial Features and Foggy Faces (2017, methodology): “For the sake of fair comparison, we report only results of methods using unconstrained types of images [16, 9, 41, 42, 43, 8, 18, 12] and we omit results of methods using only frontal or near-frontal face images.”
Deep Expectation of Real and Apparent Age from a Single Image Without Facial Landmarks (2016, methodology): “Table 6 Age group estimation results [mean accuracy(%)± standard deviation] on Adience benchmark (Eidinger et al. 2014)”
Understanding and Comparing Deep Neural Networks for Age and Gender Classification (2017, methodology): “One of the more recent face image data sets is the Adience benchmark [8], which has been published in 2014, containing 26,580 photos across 2,284 subjects with a binary gender label and one label from eight different age groups(1), partitioned into five splits.”
Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications (2021, methodology): “First, test results after training using stochastic gradient descent (SGD) [108] report reasonable performance, with exact and one-off [44], [168] prediction accuracy(3) of 56.”
ChaLearn Looking at People and Faces of the World: Face AnalysisWorkshop and Challenge 2016 (2016, neutral): “OUI-Adience [8] 26.”
NDDR-CNN: Layerwise Feature Fusing in Multi-Task CNNs by Neural Discriminative Dimensionality Reduction (2018, methodology): “We also show our performance for the age and gender classification on the Adience-Faces dataset [6] in Table S4 in the Supplementary Materials, where we compare with more state-ofthe-art methods reported on this dataset, such as [53].”
DAGER: Deep Age, Gender and Emotion Recognition Using Convolutional Neural Network (2017, methodology): “Table 6: Results for gender recognition on the Adience benchmark [11].”
Learning Not to Learn: Training Deep Neural Networks With Biased Data (2018, methodology): “To filter out misannotated images, we used pretrained networks [15] on Adience benchmark [7] designed for age and gender classification.”
Sentiment analysis in organizational work: Towards an ontology of people analytics (2018, methodology): “Such studies include: Assessing age and gender based on a picture (Tian & Chen, 2017; Geng, Zhou & Smith-Miles, 2007; Eidinger, Enbar & Hassner, 2014); Sentiment identification by body motions (Stock, Righart & Gelder, 2007) and tone of voice; Activity profiling through body motion analysis (Preece…”
Age and gender classification using convolutional neural networks (2015, methodology): “Method Accuracy Best from [10] 77.8 ± 1.3 Best from [23] Table 2.”