To compare different object detection models, we need common evaluation metrics. For a given test set, we run each model and gather its predictions. We use the predictions and the ground truth to compute an evaluation metric. In this section, we will have a look at the metrics used to evaluate object detection models.





















































