Table 7

A summary of the currently proposed evaluation metrics.

MetricAdvantageDisadvantage
F1-scoreFocuses on evaluating the presence of a predicted object corresponding to the ground truth objectDoes not account for pixel-level errors
IoU (Jaccard Index)Measures the conformance of shape between ground truth and predictionDoes not account for object level errors
Dice Similarity Coefficient (DSC)Measure of pixel wise agreement between ground truth and predictionDoes not penalize detection errors
Aggregated Jaccard Index (AJI)Penalizes both object level and pixel level errorsOver penalization owing to failed detections
Panoptic Quality (PQ)Unified scoring of detection and segmentationDependent on IoU with a strict threshold and hence may result in a lower score

or Create an Account

Close subscription notice
Close access options