ImpRoving Conditional Generative Adversarial Networks for Inverse Design of Plasmonic Nanostructures Using Label Projection and Embedding Methods
We introduce label projection and embedding methods to improve generative adversarial networks. As proof of concept, we generate images of plasmonic nanostructures with different shapes by using a model trained on their linear optical response.
Authors
- Nicolò Maccaferri (ORCID: https://orcid.org/0000-0002-0143-1510)
- Nils Henriksson (ORCID: https://orcid.org/0000-0003-0992-9871)
- P. A. Persson
Institutions
- Umeå University (SE)
Publication Details
- Published
- 2025-06-23
- DOI
- https://doi.org/10.1109/cleo/europe-eqec65582.2025.11110418
- Citations
- 1
- Primary Topic
- Image Enhancement Techniques
- Type
- article
- Field-Weighted Citation Impact
- 1.28