Gender recognition from unconstrained selfie images: a convolutional neural network approach

<p>Human gender recognition is an essential demographic tool. This is reflected in forensic science, surveillance systems and targeted marketing applications. This research was always driven using standard face images and hand-crafted features. Such way has achieved good results, however, the reliability of the facial images had a great effect on the robustness of extracted features, where any small change in the query facial image could change the results. Nevertheless, the performance of current techniques in unconstrained environments is still inefficient, especially when contrasted against recent breakthroughs in different computer vision research. This paper introduces a novel technique for human gender recognition from non-standard selfie images using deep learning approaches. Selfie photos are uncontrolled partial or full-frontal body images that are usually taken by people themselves in real-life environment. As far as we know this is the first paper of its kind to identify gender from selfie photos, using deep learning approach. The experimental results on the selfie dataset emphasizes the proposed technique effectiveness in recognizing gender from such images with 89% accuracy. The performance is further consolidated by testing on numerous benchmark datasets that are widely used in the field, namely: Adience, LFW, FERET, NIVE, Caltech WebFaces and<br />CAS-PEAL-R1.</p>

Age and Gender Prediction using Face Recognition

Automatic age and gender prediction from face images has lately attracted much attention due to its wide range of applications in numerous facial analyses. We show in this study that utilizing the Caffe Model Architecture of Deep Learning Frame Work; we were able to greatly enhance age and gender recognition by learning representations using deep-convolutional neural networks (CNN). We propose a much simpler convolutional net architecture that can be employed even if no learning data is available. In a recent study presenting a potential benchmark for age and gender estimation, we show that our strategy greatly outperforms existing state-of-the-art methods.

The Use of QLRBP and MLLPQ as Feature Extractors Combined with SVM and kNN Classifiers for Gender Recognition

Security systems must be continuously developed in order to cope with new challenges. One example of such challenges is the proliferation of sexual harassment against women in public places, such as public toilets and public transportation. Although separately designated toilets or waiting and seating areas in public transports are provided, enforcing these restrictions need constant manual surveillance. In this paper we propose an automatic gender classification system based on an individual’s facial characteristics. We evaluate the performance of QLRBP and MLLPQ as feature extractors combined with SVM or kNN as classifiers. Our experiments show that MLLPQ gives superior performance compared to QLRBP for either classifier. Furthermore, MLLPQ is less computationally demanding compared to QLRBP. The best result we achieved in our experiments was the combination of MLLPQ and kNN classifier, yielding an accuracy rate of 92.11%.

MEDICINSKE USLUGE POLNOG PRILAGOĐAVANjA KAO USLOV ZA PRAVNU PROMENU POLA U SVETLU PRAKSE EVROPSKOG SUDA ZA LjUDSKA PRAVA

By its decision in case Goodwin v. United Kingdom (2002), The European Court of Human Rights has recognized the positive obligation of states to provide conditions for the legal recognition of preferred gender in the context of the right to respect for private life. In this regard, the Court emphasized gender identity as an important element of personal identity and an integral part of the transgender person's right to private life. On the other hand, states have kept their margin of appreciation regarding requirements needed for changing gender data in civil registries or in other words legal recognition of preferred gender. After Goodwin case, that has laid foundations for the rights of transgender people to gender identity, further development of this right was set by the decision of the European Court of Human Rights in case A.P., Garçon and Nicot v. France (2017). By this decision, the Court has further narrow the margin of appreciation removing imposing of sterilisation as a requirement for legal gender recognition. Finally, The European Court of Human Rights has taken the position in the latest judgment X and Y. v. Romania (2021) that conditioning legal recognition of preferred gender with surgical interventions of gender reassignment represents breach of the right to respect private life. Thus, the Court further approached Council of Europe Resolution 1728 (2010) according to which states are suggested to remove from the requirements for legal gender recognition the subjection to any medical service of gender reassignment, including hormone therapy. Domestic legislation has retained only hormone therapy as a necessary condition for legal gender reassignment. Although this solution is in accordance with the latest case law of the European Court of Human Rights, another step is needed to make the exercise of the right to gender identity adjusted to the "soft law" of the Council of Europe and the bodies under the auspices of the United Nations.

A Comparison of Deep Learning Architectures for Automatic Gender Recognition from Audio Signals

O reconhecimento de gênero a partir da fala é um problema relacionado à análise de fala humana, e possui diversas aplicações que vão desde a personalização na recomendação de produtos à ciência forense. A identificação da eficiência e custos de diferentes abordagens que lidam com esse problema é imprescindível. Este trabalho tem como foco investigar e comparar a eficiência e custos de diferentes arquiteturas de deep learning para o reconhecimento de gênero a partir da fala. Os resultados mostram que o modelo convolucional unidimensional consegue os melhores resultados. No entanto, constatou-se que o modelo fully connected apresentou resultados próximos com menor custo, tanto no uso de memória, quanto no tempo de treinamento.

The Gender Recognition Act: Past, Present and Future

This article shall focus on the landmark 2004 Gender Recognition Act and associated legal cases. It will explore the legal rulings that lead to the Act being passed, the content of the Act and the impact this had on the transgender community in the UK, including subsequent legal issues.