Robust gender recognition by exploiting facial attributes dependencies

Bekios-Calfa, Juan; Buenaposada, José Miguel; Baumela, Luis

doi:10.1016/j.patrec.2013.04.028

Cited by 82 publications

(46 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…testing with an independent dataset, whose images were captured with different conditions. Crossdatabase classification is closer to real situations with no dataset bias, that has been proven to provoke optimistic accuracies [7,19]. Indeed, in real scenarios a gender classifier is trained with a set of images, and later deployed under conditions that may differ from those of the training dataset.…”

Section: Related Workmentioning

confidence: 99%

Descriptors and regions of interest fusion for in- and cross-database gender classification in the wild

Castrilln-Santana

Lorenzo-Navarro

Ramn-Balmaseda

2017

Image and Vision Computing

View full text Add to dashboard Cite

Gender classification (GC) has achieved high accuracy in different experimental evaluations based mostly on inner facial details. However, these results do not generalize well in unrestricted datasets and particularly in cross-database experiments, where the performance drops drastically. In this paper, we analyze the state-of-the-art GC accuracy on three large datasets: MORPH, LFW and GROUPS. We discuss their respective difficulties and bias, concluding that the most challenging and wildest complexity is present in GROUPS. This dataset covers hard conditions such as low resolution imagery and cluttered background. Firstly, we analyze in depth the performance of different descriptors extracted from the face and its local context on this dataset. Selecting the bests and studying their most suitable combination allows us to design a solution that beats any previously published results for GROUPS with the Dago's protocol, reaching an accuracy over 94.2%, reducing the gap with other simpler datasets. The chosen solution based on local descriptors is later evaluated in a cross-database scenario with the three mentioned datasets, and full dataset 5-fold cross validation. The achieved results are compared with a Convolutional Neural Network approach, achieving rather similar marks. Finally, a solution is proposed combining both focuses, exhibiting great complementarity, boosting GC performance to beat previously published results in GC both cross-database, and full in-database evaluations.

show abstract

Section: Related Workmentioning

confidence: 99%

Descriptors and regions of interest fusion for in- and cross-database gender classification in the wild

Castrilln-Santana

Lorenzo-Navarro

Ramn-Balmaseda

2017

Image and Vision Computing

View full text Add to dashboard Cite

show abstract

“…Sekara and Lehmann [29] studied how to deploy the electronic datasets as a valid proxy for real life social interactions, and found that the strength of electronic signals can be used to distinguish between strong and weak friendship ties. Some other information has been used to predict users' demographics: predicting authors' age and gender from their writing styles [30]; gender classification from videos analysis [31]; and gender/age estimation from images of faces [32,33]. Due to the scope of this study, we will not detail existing research in this direction.…”

Section: Social Behaviormentioning

confidence: 99%

Understanding the predictability of user demographics from cyber-physical-social behaviours in indoor retail spaces

et al. 2018

View full text Add to dashboard Cite

Understanding the association between customer demographics and behaviour is critical for operators of indoor retail spaces. This study explores such an association based on a combined understanding of customer Cyber (online), Physical, and (some aspects of ) Social (CPS) behaviour, at the conjunction of corresponding CPS spaces. We combine the results of a traditional questionnaire with large-scale WiFi access logs, which capture customer cyber and physical behaviour. We investigate the predictability of user demographics based on CPS behaviors captured from both sources. We find (1) strong correlations between users' demographics and their CPS behaviors; (2) log-recorded cyber-physical behavior reflects well data captured in the corresponding questionnaire; (3) different CPS behaviors contribute differently to the predictability of demographic attributes; and (4) the predictability of user demographics from logs is comparable to questionnaire-based data. As such, our study provides strong support for demographic studies based on large-scale logs data capture.

show abstract

“…Juan Bekios-Calfa, Jose m.Buenaposada and Luis Baumela [8], studied the problem of gender recognition from a multiattribute perspective. Gender recognition under constrained conditions (e.g.…”

Section: Figure 1 Typical Framework For Gender Classification Systemmentioning

confidence: 99%

An Analysis on Gender Classification and Age Estimation Approaches

Kaur¹,

Rai²

2017

IJCA

View full text Add to dashboard Cite

There has been a growing interest in automatic age and gender classification, as it has become relevant to an increasing amount of applications such as human-computer interaction, surveillance, biometrics, intelligent marketing and many more. Facial age and gender from the face image of a person is one such significant demographic attribute. In this paper, presents a review of automatic facial gender classification and age estimation framework in computer vision. While highlighting the challenges involved during classification of images captured under unconstrained conditions or may be the laborious process of gathering the face images for age estimation, as aging is the uncontrolled and slow process. A comprehensive survey for facial feature extraction methods and face databases for gender and age estimation studied in the past couple of decades is mentioned. Evaluation and result based performance achieved for various face images from different databases has been explained.

show abstract

Robust gender recognition by exploiting facial attributes dependencies

Cited by 82 publications

References 24 publications

Descriptors and regions of interest fusion for in- and cross-database gender classification in the wild

Descriptors and regions of interest fusion for in- and cross-database gender classification in the wild

Understanding the predictability of user demographics from cyber-physical-social behaviours in indoor retail spaces

An Analysis on Gender Classification and Age Estimation Approaches

Contact Info

Product

Resources

About