Survey on Sketch-to-photo Translation
ACM Computing Surveys · 31.01.2024 · 56 (1) · 1-25
DOI: 10.1145/3606694 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Computing Surveys, online erschienen am 26.08.2023. Verfasst von Diego Donoso und 1 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 51 Literaturangaben. Gefördert von: National Center for Artificial Intelligence.
Abstract (Originalsprache)
Sketch-based understanding is involved in human communication and cognitive development, making it essential in visual perception. A specific task in this domain is sketch-to-photo translation, where a model produces realistic images from simple drawings. To this end, large paired training datasets are commonly required, which is impractical in real applications. Thus, this work studies conditional generative models for sketch-to-photo translation, overcoming the lack of training datasets by a self-supervised approach that produces sketch-photo pairs from a target catalog. Our study shows the benefit of cycle-consistency loss and UNet architectures that, together with the proposed dataset generation, improve performance in real applications like eCommerce. Our results also reveal the weakness of conditional DDPMs for generating images resembling the input sketch, even though they achieve a high FID score.
Förderung
- National Center for Artificial Intelligence
Zitieren
APA
Donoso, D., & Saavedra, J. M. (2024). Survey on Sketch-to-photo Translation. ACM Computing Surveys, 56(1), 1-25. https://doi.org/10.1145/3606694
Vancouver
Donoso D, Saavedra JM. Survey on Sketch-to-photo Translation. ACM Computing Surveys. 2024;56(1):1-25. doi:10.1145/3606694
BibTeX
@article{Donoso2024,
author = {Donoso, Diego and Saavedra, Jose M.},
title = {Survey on Sketch-to-photo Translation},
journal = {ACM Computing Surveys},
year = {2024},
volume = {56},
number = {1},
pages = {1--25},
doi = {10.1145/3606694}
}
EUVIMED