A Semi-Automated Usability Evaluation Framework for Interactive Image Segmentation Systems

Table 4

The five most important features per GBRF estimator/label. Italic indicates the most frequently used features in the trained decision trees of the GBRFs. Bold highlights semantically similar feature pairs. The abbreviations represent the receiver operating characteristic area under the curve (ROC_AUC), logistic loss (LOG), and relative absolute area/volume difference (RAVD).

	1.	2.	3.	4.	5.
ATT	Mean(ROC_AUC/wtime)	Mean(Dice)/Mean(wtime)	Mean(LOG)/Mean(ctime)	Med(OBJ_TPR)/Med(ctime)	Med(ctime)
HQ-I	Mean(ROC_AUC/wtime)	PCA_VAL_17	Mean(Dice)/Mean(wtime)	Med(Med_ctime)/Med(wtime)	Mean(LOG)/Mean(ctime)
HQ	Med(Jaccard/ctime)	PCA_VAL_17	Mean(ROC_AUC/wtime)	Mean(OBJ_TPR/wtime)	Mean(RAVD/ctime)
HQ-S	Mean(RAVD)/Mean(ctime)	Med(Med_wtime/wtime)	Med(LOG)	Std(Relative_Seed_Coord_H)	Med(MSE)
PQ	PCA_VAL_16	Mean(otime/ctime)	Mean(Dice)/Mean(ctime)	PCA_VAL_11	Med(Med_ctime/wtime)
SUS	PCA_VAL_2	PCA_VAL_18	Std(Relative_Seed_Coord_H)	Med(Med_wtime)	PCA_VAL_20