Solano Carrillo, Edgardo und Bueno Rodriguez, Angel und Carrillo Perez, Borja Jesus und Steiniger, Yannik und Stoppe, Jannis (2023) Look ATME: The Discriminator Mean Entropy Needs Attention. In: 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023. CVPR 2023 Workshop on Generative Models in Computer Vision, 2023-06-17 - 2023-06-24, Vancouver, Canada. doi: 10.1109/CVPRW59228.2023.00086. ISBN 979-835030129-8. ISSN 1063-6919.
PDF
5MB |
Kurzfassung
Generative adversarial networks (GANs) are successfully used for image synthesis but are known to face instability during training. In contrast, probabilistic diffusion models (DMs) are stable and generate high-quality images, at the cost of an expensive sampling procedure. In this paper, we introduce a simple method to allow GANs to stably converge to their theoretical optimum, while bringing in the denoising machinery from DMs. These models are combined into a simpler model (ATME) that only requires a forward pass during inference, making predictions cheaper and more accurate than DMs and popular GANs. ATME breaks an information asymmetry existing in most GAN models in which the discriminator has spatial knowledge of where the generator is failing. To restore the information symmetry, the generator is endowed with knowledge of the entropic state of the discriminator, which is leveraged to allow the adversarial game to converge towards equilibrium. We demonstrate the power of our method in several image-to-image translation tasks, showing superior performance than state-of-the-art methods at a lesser cost. Code is available at https://github.com/DLR-MI/atme.
elib-URL des Eintrags: | https://elib.dlr.de/195910/ | ||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Dokumentart: | Konferenzbeitrag (Poster) | ||||||||||||||||||||||||
Titel: | Look ATME: The Discriminator Mean Entropy Needs Attention | ||||||||||||||||||||||||
Autoren: |
| ||||||||||||||||||||||||
Datum: | 2023 | ||||||||||||||||||||||||
Erschienen in: | 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023 | ||||||||||||||||||||||||
Referierte Publikation: | Ja | ||||||||||||||||||||||||
Open Access: | Ja | ||||||||||||||||||||||||
Gold Open Access: | Nein | ||||||||||||||||||||||||
In SCOPUS: | Ja | ||||||||||||||||||||||||
In ISI Web of Science: | Ja | ||||||||||||||||||||||||
DOI: | 10.1109/CVPRW59228.2023.00086 | ||||||||||||||||||||||||
ISSN: | 1063-6919 | ||||||||||||||||||||||||
ISBN: | 979-835030129-8 | ||||||||||||||||||||||||
Status: | veröffentlicht | ||||||||||||||||||||||||
Stichwörter: | Deep learning, GAN, pix2pix, image-to-image translation, diffusion models | ||||||||||||||||||||||||
Veranstaltungstitel: | CVPR 2023 Workshop on Generative Models in Computer Vision | ||||||||||||||||||||||||
Veranstaltungsort: | Vancouver, Canada | ||||||||||||||||||||||||
Veranstaltungsart: | internationale Konferenz | ||||||||||||||||||||||||
Veranstaltungsbeginn: | 17 Juni 2023 | ||||||||||||||||||||||||
Veranstaltungsende: | 24 Juni 2023 | ||||||||||||||||||||||||
HGF - Forschungsbereich: | Luftfahrt, Raumfahrt und Verkehr | ||||||||||||||||||||||||
HGF - Programm: | Verkehr | ||||||||||||||||||||||||
HGF - Programmthema: | keine Zuordnung | ||||||||||||||||||||||||
DLR - Schwerpunkt: | Verkehr | ||||||||||||||||||||||||
DLR - Forschungsgebiet: | V - keine Zuordnung | ||||||||||||||||||||||||
DLR - Teilgebiet (Projekt, Vorhaben): | V - keine Zuordnung | ||||||||||||||||||||||||
Standort: | Bremerhaven | ||||||||||||||||||||||||
Institute & Einrichtungen: | Institut für den Schutz maritimer Infrastrukturen | ||||||||||||||||||||||||
Hinterlegt von: | Solano Carrillo, Edgardo | ||||||||||||||||||||||||
Hinterlegt am: | 20 Nov 2023 14:00 | ||||||||||||||||||||||||
Letzte Änderung: | 12 Jul 2024 08:23 |
Nur für Mitarbeiter des Archivs: Kontrollseite des Eintrags