Sensitivity of Average Precision to Bounding Box Perturbations

06/21/2022

∙

by   Ali Borji, et al.

∙

0

∙

share

Object detection is a fundamental vision task. It has been highly researched in academia and has been widely adopted in industry. Average Precision (AP) is the standard score for evaluating object detectors. Our understanding of the subtleties of this score, however, is limited. Here, we quantify the sensitivity of AP to bounding box perturbations and show that AP is very sensitive to small translations. Only one pixel shift is enough to drop the mAP of a model by 8.4 is 23.1 predictions are 23 achieving higher mAP becomes increasingly harder as models get better. We also investigate the effect of box scaling on AP. Code and data is available at https://github.com/aliborji/AP_Box_Perturbation.