gdiprisco/Multitask_CNNs_for_efficient_face_analysis_in_the_wild

An innovative multi-task neural architecture suitable for face analysis and the consequent identification of gender, age, ethnicity and emotion. A network capable of the performance of the state of the art of single-task networks by sharing the feature extraction part. It may be suitable for robotic applications: it requires time and memory in the order of a single-task network.

Python