Conceptual Cognitive Modeling for Fine-Grained Annotation Quality Assessment of Object Detection Datasets

Table 3

Annotation accuracy of ground truth annotations for the downsampled UTS dataset.

Class	Flat accuracy of label		Hierarchical accuracy of label
	Precision	Recall	Precision	Recall
Hatchback	0.57 ± 0.13	0.57 ± 0.04	0.85 ± 0.06	0.67 ± 0.05
Sedan	0.79 ± 0.12	0.44 ± 0.04	0.91 ± 0.08	0.65 ± 0.02
Minibus	0.58 ± 0.34	0.77 ± 0.17	0.72 ± 0.23	0.79 ± 0.17
SUV	0.27 ± 0.10	0.15 ± 0.05	0.75 ± 0.04	0.53 ± 0.09
MPV	0.26 ± 0.14	0.17 ± 0.15	0.72 ± 0.08	0.51 ± 0.18
Pickup	1 ± 0	0.28 ± 0.19	1 ± 0	0.28 ± 0.19
On average	0.58 ± 0.07	0.40 ± 0.07	0.83 ± 0.04	0.57 ± 0.07