Face Detection with RetinaNet - work in progress

In this notebook, the goal is train a RetinaNet model with a ResNet-50 backbone on an extract from the FDDB dataset (Face Detection Data Set and Benchmark). RetinaNet is a popular single-stage detector; It uses a feature pyramid network to efficiently detect objects at multiple scales and introduces a new loss, the Focal loss function, to alleviate the problem of the extreme foreground-background class imbalance.