A Comprehensive Evaluation Framework for the Study of the Effects of Facial Filters on Face Recognition Accuracy
This work addresses the need for larger-scale evaluation of facial filters on automated face recognition for social media applications, but it is incremental as it builds on prior studies with a more systematic approach.
The authors tackled the problem of limited studies on facial filters' impact on face recognition by introducing a comprehensive evaluation framework, which includes a controlled dataset, principled filter selection, and experiments, and demonstrated it with a case study on cross-cultural differences and a method to detect and restore filtering effects to improve performance.
Facial filters are now commonplace for social media users around the world. Previous work has demonstrated that facial filters can negatively impact automated face recognition performance. However, these studies focus on small numbers of hand-picked filters in particular styles. In order to more effectively incorporate the wide ranges of filters present on various social media applications, we introduce a framework that allows for larger-scale study of the impact of facial filters on automated recognition. This framework includes a controlled dataset of face images, a principled filter selection process that selects a representative range of filters for experimentation, and a set of experiments to evaluate the filters' impact on recognition. We demonstrate our framework with a case study of filters from the American applications Instagram and Snapchat and the Chinese applications Meitu and Pitu to uncover cross-cultural differences. Finally, we show how the filtering effect in a face embedding space can easily be detected and restored to improve face recognition performance.