NoMoColor: Unified Noise Modulation for Enhanced Diffusion-based Image Colorization (Student Abstract)
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (48) · 41182-41184
DOI: 10.1609/aaai.v40i48.42207 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Ankan Deria und 3 weiteren Autor:innen.
Abstract (Originalsprache)
We present a language-based noise modulation module for diffusion models that improves image color generation under textual guidance. Unlike standard approaches that inject noise uniformly, our method leverages semantic cues from text to selectively control the noise injection process, preserving local details and enhancing color accuracy even when descriptions are ambiguous or incomplete. Applied to language guided image colorization, this targeted modulation leads to more faithful and visually consistent results. The proposed module is lightweight, generalizable, and can be integrated into existing diffusion pipelines, offering a simple yet effective step toward more controllable text-to-image generation.
Zitieren
APA
Deria, A., Mahapatra, D., Mondal, M., & Roy, S. (2026). NoMoColor: Unified Noise Modulation for Enhanced Diffusion-based Image Colorization (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence, 40(48), 41182-41184. https://doi.org/10.1609/aaai.v40i48.42207
Vancouver
Deria A, Mahapatra D, Mondal M, Roy S. NoMoColor: Unified Noise Modulation for Enhanced Diffusion-based Image Colorization (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(48):41182-41184. doi:10.1609/aaai.v40i48.42207
BibTeX
@article{Deria2026,
author = {Deria, Ankan and Mahapatra, Dwarikanath and Mondal, Murari and Roy, Sudipta},
title = {NoMoColor: Unified Noise Modulation for Enhanced Diffusion-based Image Colorization (Student Abstract)},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {48},
pages = {41182--41184},
doi = {10.1609/aaai.v40i48.42207}
}
EUVIMED