Deep learning object detection as an assistance system for complex image labeling tasks

Abstract

Object detection via deep learning has many promising areas of application. However, robustness and accuracy of fully automated systems are often insufficient for practical use. Integrating results from Artificial Intelligence (AI) and human intelligence in collaborative settings might bridge the gap between efficiency and accuracy. This study proves increased efficiency when supporting human intelligence through AI without negative impact on effectiveness in a finegrained car scratch image labeling task. Based on the confirmed benefits of AI with human intelligence in the loop approaches, this contribution discusses potential practical application scenarios and envisions the implementation of assistance systems supported by computer vision.

Understanding online fake review production strategies

“…Despite ongoing efforts, current detection algorithms have yet to be proven effective in practice (Luca & Zervas, 2016;Reyes-Menendez et al, 2019). This is because NLG-based bots, which are often used in computational experiments, tend to produce largely consistent patterns when compared with human spammers who are more intuitive and less predictable (Leimkühler et al, 2021). Moreover, as fake reviewers grow in sophistication, it becomes harder to weed out phony entries algorithmically.…”