Reliable detection of doppelgängers based on deep face representations
Abstract Doppelgängers (or lookalikes) usually yield an increased probability of false matches in a facial recognition system, as opposed to random face image pairs selected for non‐mated comparison trials. In this work, the impact of doppelgängers on the HDA Doppelgänger and Disguised Faces in The...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-05-01
|
| Series: | IET Biometrics |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/bme2.12072 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832559720971370496 |
|---|---|
| author | Christian Rathgeb Daniel Fischer Pawel Drozdowski Christoph Busch |
| author_facet | Christian Rathgeb Daniel Fischer Pawel Drozdowski Christoph Busch |
| author_sort | Christian Rathgeb |
| collection | DOAJ |
| description | Abstract Doppelgängers (or lookalikes) usually yield an increased probability of false matches in a facial recognition system, as opposed to random face image pairs selected for non‐mated comparison trials. In this work, the impact of doppelgängers on the HDA Doppelgänger and Disguised Faces in The Wild databases is assessed using a state‐of‐the‐art face recognition system. It is found that doppelgänger image pairs yield very high similarity scores resulting in a significant increase of false match rates. Further, a doppelgänger detection method is proposed, which distinguishes doppelgängers from mated comparison trials by analysing differences in deep representations obtained from face image pairs. The proposed detection system employs a machine learning‐based classifier, which is trained with generated doppelgänger image pairs utilising face morphing techniques. Experimental evaluations conducted on the HDA Doppelgänger and Look‐Alike Face databases reveal a detection equal error rate of approximately 2.7% for the task of separating mated authentication attempts from doppelgängers. |
| format | Article |
| id | doaj-art-c50d412417054ff487833f96cb4e9d50 |
| institution | Kabale University |
| issn | 2047-4938 2047-4946 |
| language | English |
| publishDate | 2022-05-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Biometrics |
| spelling | doaj-art-c50d412417054ff487833f96cb4e9d502025-02-03T01:29:24ZengWileyIET Biometrics2047-49382047-49462022-05-0111321522410.1049/bme2.12072Reliable detection of doppelgängers based on deep face representationsChristian Rathgeb0Daniel Fischer1Pawel Drozdowski2Christoph Busch3da/sec ‐ Biometrics and Internet Security Research Group Hochschule Darmstadt Darmstadt Germanyda/sec ‐ Biometrics and Internet Security Research Group Hochschule Darmstadt Darmstadt Germanyda/sec ‐ Biometrics and Internet Security Research Group Hochschule Darmstadt Darmstadt Germanyda/sec ‐ Biometrics and Internet Security Research Group Hochschule Darmstadt Darmstadt GermanyAbstract Doppelgängers (or lookalikes) usually yield an increased probability of false matches in a facial recognition system, as opposed to random face image pairs selected for non‐mated comparison trials. In this work, the impact of doppelgängers on the HDA Doppelgänger and Disguised Faces in The Wild databases is assessed using a state‐of‐the‐art face recognition system. It is found that doppelgänger image pairs yield very high similarity scores resulting in a significant increase of false match rates. Further, a doppelgänger detection method is proposed, which distinguishes doppelgängers from mated comparison trials by analysing differences in deep representations obtained from face image pairs. The proposed detection system employs a machine learning‐based classifier, which is trained with generated doppelgänger image pairs utilising face morphing techniques. Experimental evaluations conducted on the HDA Doppelgänger and Look‐Alike Face databases reveal a detection equal error rate of approximately 2.7% for the task of separating mated authentication attempts from doppelgängers.https://doi.org/10.1049/bme2.12072biometricsdatabasedetectiondoppelgängerface recognitionlookalike |
| spellingShingle | Christian Rathgeb Daniel Fischer Pawel Drozdowski Christoph Busch Reliable detection of doppelgängers based on deep face representations IET Biometrics biometrics database detection doppelgänger face recognition lookalike |
| title | Reliable detection of doppelgängers based on deep face representations |
| title_full | Reliable detection of doppelgängers based on deep face representations |
| title_fullStr | Reliable detection of doppelgängers based on deep face representations |
| title_full_unstemmed | Reliable detection of doppelgängers based on deep face representations |
| title_short | Reliable detection of doppelgängers based on deep face representations |
| title_sort | reliable detection of doppelgangers based on deep face representations |
| topic | biometrics database detection doppelgänger face recognition lookalike |
| url | https://doi.org/10.1049/bme2.12072 |
| work_keys_str_mv | AT christianrathgeb reliabledetectionofdoppelgangersbasedondeepfacerepresentations AT danielfischer reliabledetectionofdoppelgangersbasedondeepfacerepresentations AT paweldrozdowski reliabledetectionofdoppelgangersbasedondeepfacerepresentations AT christophbusch reliabledetectionofdoppelgangersbasedondeepfacerepresentations |