Abstract
Facial expression recognition (FER) is a topic attracting significant research in both psychology and machine learning with a wide range of applications. Despite a wealth of research on human FER and considerable progress in computational FER made possible by deep neural networks (DNNs), comparatively less work has been done on comparing the degree to which DNNs may be comparable to human performance. In this work, we compared the recognition performance and attention patterns of humans and machines during a two-alternative forced-choice FER task. Human attention was here gathered through click data that progressively uncovered a face, whereas model attention was obtained using three different popular techniques from explainable AI: CAM, GradCAM and Extremal Perturbation. In both cases, performance was gathered as percent correct. For this task, we found that humans outperformed machines quite significantly. In terms of attention patterns, we found that Extremal Perturbation had the best overall fit with the human attention map during the task.
Original language | English |
---|---|
Title of host publication | Pattern Recognition - 6th Asian Conference, ACPR 2021, Revised Selected Papers |
Editors | Christian Wallraven, Qingshan Liu, Hajime Nagahara |
Publisher | Springer Science and Business Media Deutschland GmbH |
Pages | 403-416 |
Number of pages | 14 |
ISBN (Print) | 9783031023743 |
DOIs | |
Publication status | Published - 2022 |
Event | 6th Asian Conference on Pattern Recognition, ACPR 2021 - Virtual, Online Duration: 2021 Nov 9 → 2021 Nov 12 |
Publication series
Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|
Volume | 13188 LNCS |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Conference
Conference | 6th Asian Conference on Pattern Recognition, ACPR 2021 |
---|---|
City | Virtual, Online |
Period | 21/11/9 → 21/11/12 |
Bibliographical note
Funding Information:Acknowledgments. This work was supported by Institute of Information Communications Technology Planning Evaluation (IITP; No. 2019-0-00079, Department of Artificial Intelligence, Korea University) and National Research Foundation of Korea (NRF; NRF-2017M3C7A1041824) grant funded by the Korean government (MSIT).
Publisher Copyright:
© 2022, Springer Nature Switzerland AG.
Keywords
- AffectNet
- Facial expression recognition
- Human-in-the-loop
- Humans versus machines
ASJC Scopus subject areas
- Theoretical Computer Science
- Computer Science(all)