Introduction to Visual Perception

We start our treatment of visual perception by defining the specific visual perception tasks and discussing how each of them can be evaluated. Namely, we introduce the problems of image classification, image segmentation, instance segmentation, and object detection. We define precision and recall and introduce the precision-recall (PR) curve.

We also define intersection-over-union (IoU) and mean average precision (mAP), which are metrics used to evaluate object detection algorithms.