Exploiting Face Recognizability with Early Exit Vision Transformers
| dc.contributor.author | Seth Nixon, Pietro Ruiu | |
| dc.contributor.editor | Damer, Naser | |
| dc.contributor.editor | Gomez-Barrero, Marta | |
| dc.contributor.editor | Raja, Kiran | |
| dc.contributor.editor | Rathgeb, Christian | |
| dc.contributor.editor | Sequeira, Ana F. | |
| dc.contributor.editor | Todisco, Massimiliano | |
| dc.contributor.editor | Uhl, Andreas | |
| dc.date.accessioned | 2023-12-12T10:46:46Z | |
| dc.date.available | 2023-12-12T10:46:46Z | |
| dc.date.issued | 2023 | |
| dc.description.abstract | Face recognition with Deep Learning is generally approached as a problem of capacity. The field has seen progressively deeper, more complex models or larger, more highly variant datasets. However, the carbon footprint of machine learning is a concern. A real push is developing to reduce the energy consumption of machine learning as we strive for a more eco-friendly society. Lower energy consumption or compute budget is always desirable, if accuracy is not reduced below a usable level. We present an approach using the state of the art Vision Transformer and Early Exits for reducing compute budget without significantly affecting performance. We develop a system for face recognition and identification with a closed-set gallery and show that with a small reduction in performance, a reasonable reduction in compute cost can be obtained using our method. | en |
| dc.identifier.isbn | 978-3-88579-733-3 | |
| dc.identifier.issn | 1617-5468 | |
| dc.identifier.uri | https://dl.gi.de/handle/20.500.12116/43261 | |
| dc.language.iso | en | |
| dc.pubPlace | Bonn | |
| dc.publisher | Gesellschaft für Informatik e.V. | |
| dc.relation.ispartof | BIOSIG 2023 | |
| dc.relation.ispartofseries | Lecture Notes in Informatics (LNI) - Proceedings, Volume P-339 | |
| dc.subject | Computational efficiency in biometrics | |
| dc.subject | Face and gesture recognition | |
| dc.title | Exploiting Face Recognizability with Early Exit Vision Transformers | en |
| dc.type | Text/Conference Paper | |
| mci.conference.date | 20.-22. September 2023 | |
| mci.conference.location | Darmstadt | |
| mci.conference.sessiontitle | Regular Research Papers | |
| mci.reference.pages | 132-143 |
Dateien
Originalbündel
1 - 1 von 1
Vorschaubild nicht verfügbar
- Name:
- LNI_045.pdf
- Größe:
- 723.82 KB
- Format:
- Adobe Portable Document Format