What is Faster R-CNN?

Faster R-CNN is an advanced object detection model based on the Region-Based Convolutional Neural Network (R-CNN) architecture. A convolutional neural network extracts features, then a region proposal network (RPN) generates candidate regions likely to contain objects, and these regions are then classified and their bounding boxes refined. Faster R-CNN is more accurate than earlier methods such as R-CNN and Fast R-CNN because it merged region proposal and detection into a single network, which made it faster and more efficient in applications that require high accuracy, such as face recognition and medical image detection.