Summary of Mask-up: Investigating Biases in Face Re-identification For Masked Faces, by Siddharth D Jaiswal et al.
Mask-up: Investigating Biases in Face Re-identification for Masked Faces
by Siddharth D Jaiswal, Ankit Kr. Verma, Animesh Mukherjee
First submitted to arxiv on: 21 Feb 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper presents a comprehensive evaluation of Face Recognition Systems (FRSs) in response to the COVID-19 pandemic and its impact on facial recognition technology. Specifically, it focuses on the accuracy and bias of FRSs when faced with masked images, a problem that has not been adequately addressed by previous studies. The authors audit 13 FRSs, including commercial and open-source solutions, using five benchmark datasets containing over 14,000 images. The results show that many FRSs are highly inaccurate, with some even perpetuating biases against non-White individuals. Furthermore, a human-in-the-loop moderation approach did not alleviate the concerns. The study concludes by emphasizing the need for developers, lawmakers, and users to re-examine the design principles of FRSs, particularly in the context of face re-identification. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Face recognition technology has become widespread since the COVID-19 pandemic. However, this technology often discriminates against marginalized groups. A major problem is that these systems are not designed to handle masked faces. In this study, researchers tested 13 different face recognition systems using many images with and without masks. They found that most of these systems did a poor job of recognizing faces when the person was wearing a mask. Some even made things worse by discriminating against certain groups of people. The results show that we need to rethink how these systems work. |
Keywords
» Artificial intelligence » Face recognition » Mask
