Search

The 2015 FRVT gender classification (GC) report evidences the problems that current approaches tackle in situations with large variations in pose, illumination, background and facial expression. The report suggests that both commercial and research solutions are hardly able to reach an accuracy over 90 % for The Images of Groups dataset, a proven scenario exhibiting unrestricted or in the wild conditions. In this paper, we focus on this challenging dataset, stepping forward in GC performance by observing: 1) recent literature results combining multiple local descriptors, and 2) the psychophysics evidences of the greater importance of the ocular and mouth areas to solve this task. We therefore make use of holistic and inner facial patches to extract features, that are later combined via a score level fusion strategy. The achieved results support the main information provided by the ocular and the mouth areas. Indeed, the combination of multiscale extracted features increases the overall accuracy to over 94 %, reducing notoriously the classification error if compared with tuned holistic and deep learning approaches. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]