Minimalistic CNN-based ensemble model for gender prediction from face images

• Obtained the record gender recognition performance of 97.31% on the LFW dataset.
• Used about 10 times fewer training images than the previous state-of-the-art.
• Only publicly available training images are used.
• The trained model is optimized in terms of running time and required memory.
• The trained model is made public for download. It can be also tested via a web demo.

Despite being extensively studied in the literature, the problem of gender recognition from face images remains difficult when dealing with unconstrained images in a cross-dataset protocol. In this work, we propose a convolutional neural network ensemble model to improve the state-of-the-art accuracy of gender recognition from face images on one of the most challenging face image datasets today, LFW (Labeled Faces in the Wild). We find that convolutional neural networks need significantly less training data to obtain the state-of-the-art performance than previously proposed methods. Furthermore, our ensemble model is deliberately designed in a way that both its memory requirements and running time are minimized. This allows us to envision a potential usage of the constructed model in embedded devices or in a cloud platform for an intensive use on massive image databases.