A Noise is Worth Diffusion Guidance

Salvato in:

Dettagli Bibliografici
Pubblicato in:	arXiv.org (Dec 5, 2024), p. n/a
Autore principale:	Ahn, Donghoon
Altri autori:	Kang, Jiwon, Lee, Sanghyun, Min, Jaewon, Kim, Minjae, Jang, Wooseok, Cho, Hyoungwon, Sayak, Paul, Kim, SeonHwa, Cha, Eunju, Jin, Kyong Hwan, Kim, Seungryong
Pubblicazione:	Cornell University Library, arXiv.org
Soggetti:	Random noise Noise generation Image quality Image reconstruction Image enhancement Image processing Noise reduction
Accesso online:	Citation/Abstract Full text outside of ProQuest
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Descrizione
Abstract:	Diffusion models excel in generating high-quality images. However, current diffusion models struggle to produce reliable images without guidance methods, such as classifier-free guidance (CFG). Are guidance methods truly necessary? Observing that noise obtained via diffusion inversion can reconstruct high-quality images without guidance, we focus on the initial noise of the denoising pipeline. By mapping Gaussian noise to `guidance-free noise', we uncover that small low-magnitude low-frequency components significantly enhance the denoising process, removing the need for guidance and thus improving both inference throughput and memory. Expanding on this, we propose \ours, a novel method that replaces guidance methods with a single refinement of the initial noise. This refined noise enables high-quality image generation without guidance, within the same diffusion pipeline. Our noise-refining model leverages efficient noise-space learning, achieving rapid convergence and strong performance with just 50K text-image pairs. We validate its effectiveness across diverse metrics and analyze how refined noise can eliminate the need for guidance. See our project page: https://cvlab-kaist.github.io/NoiseRefine/.
ISSN:	2331-8422
Fonte:	Engineering Database