Implementing RoI Pooling in TensorFlow Keras

In this post we explain the basic concept and general usage of RoI (Region of Interest) pooling and provide an implementation using Keras layers and the TensorFlow backend. The intended audience for this post are people familiar with the basic theory of (Convolutional) Neural Networks and who are capable of building and running simple models using Keras. If you are here just for the code, serve yourself from this gist and do not forget to like and share the article! RoI Pooling was proposed by Ross Girshick in the Fast R-CNN paper as part of his object recognition pipeline. In the general use case for RoI Pooling we have an image-like object, and multiple regions of interest specified via bounding boxes.