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

<table class="table-group" id="tab8"><tr><td><table class="table"><tr><td class="thead-hr" colspan="3"><hr/></td></tr><tr class="thead"><td class="align_left">Annotation quality dimension</td><td class="align_center">Human annotations</td><td class="align_center">Ground truth annotations</td></tr><tr><td class="thead-hr" colspan="3"><hr/></td></tr><tr><td class="align_left">Completeness of bounding box’s quantity</td><td class="align_center">0.90 ± 0.04</td><td class="align_center">0.88 ± 0.05</td></tr><tr><td class="align_left">Completeness of bounding box’s size</td><td class="align_center">0.84 ± 0.02</td><td class="align_center">0.85</td></tr><tr><td class="align_left">Completeness of bounding box’s label</td><td class="align_center">0.9991 ± 0.0008</td><td class="align_center">1</td></tr><tr class="table-tr"><td colspan="3"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

<div>Results of other quality dimensions for the selected images and its original training dataset of the PASCAL VOC 2007 dataset.</div>

Discrete Dynamics in Nature and Society

tab8

Table 8

Table 8: Conceptual Cognitive Modeling for Fine-Grained Annotation Quality Assessment of Object Detection Datasets 