Do Object Detection Localization Errors Affect Human Performance and Trust?

Do Object Detection Localization Errors Affect Human Performance and Trust?: An Observer Performance Study

de Witte, Sven (TU Delft Electrical Engineering, Mathematics and Computer Science; TU Delft Pattern Recognition and Bioinformatics)

van Gemert, J.C. (mentor)
Strafforello, O. (mentor)
Martinez, Jorge (graduation committee)

Delft University of Technology

Computer Science

2023-12-21

Bounding boxes are often used to communicate automatic object detection results to humans, aiding humans in a multitude of tasks. We investigate the relationship between bounding box localization errors and human task performance. We use observer performance studies on a visual multi-object counting task to measure both human trust and performance with different levels of bounding box accuracy. The results show that localization errors have no significant impact on human accuracy or trust in the system. Recall and precision errors impact both human performance and trust, suggesting that optimizing algorithms based on the F1 score is more beneficial in human-computer tasks. Lastly, the paper offers an improvement on bounding boxes in multi-object counting tasks with center dots, showing improved performance and better resilience to localization
inaccuracy.

Observer Performance
Object Detection
Bounding Box
Trust

http://resolver.tudelft.nl/uuid:d7f02e67-7d9e-484d-bfc2-2cfc069bc35f

Student theses

master thesis

© 2023 Sven de Witte