Applying Guidance in a Limited Interval Improves Sample and Distribution Quality in Diffusion Models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913572313890816 |
|---|---|
| author | Kynkäänniemi, Tuomas Aittala, Miika Karras, Tero Laine, Samuli Aila, Timo Lehtinen, Jaakko |
| author_facet | Kynkäänniemi, Tuomas Aittala, Miika Karras, Tero Laine, Samuli Aila, Timo Lehtinen, Jaakko |
| contents | Guidance is a crucial technique for extracting the best performance out of image-generating diffusion models. Traditionally, a constant guidance weight has been applied throughout the sampling chain of an image. We show that guidance is clearly harmful toward the beginning of the chain (high noise levels), largely unnecessary toward the end (low noise levels), and only beneficial in the middle. We thus restrict it to a specific range of noise levels, improving both the inference speed and result quality. This limited guidance interval improves the record FID in ImageNet-512 significantly, from 1.81 to 1.40. We show that it is quantitatively and qualitatively beneficial across different sampler parameters, network architectures, and datasets, including the large-scale setting of Stable Diffusion XL. We thus suggest exposing the guidance interval as a hyperparameter in all diffusion models that use guidance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_07724 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Applying Guidance in a Limited Interval Improves Sample and Distribution Quality in Diffusion Models Kynkäänniemi, Tuomas Aittala, Miika Karras, Tero Laine, Samuli Aila, Timo Lehtinen, Jaakko Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Neural and Evolutionary Computing Guidance is a crucial technique for extracting the best performance out of image-generating diffusion models. Traditionally, a constant guidance weight has been applied throughout the sampling chain of an image. We show that guidance is clearly harmful toward the beginning of the chain (high noise levels), largely unnecessary toward the end (low noise levels), and only beneficial in the middle. We thus restrict it to a specific range of noise levels, improving both the inference speed and result quality. This limited guidance interval improves the record FID in ImageNet-512 significantly, from 1.81 to 1.40. We show that it is quantitatively and qualitatively beneficial across different sampler parameters, network architectures, and datasets, including the large-scale setting of Stable Diffusion XL. We thus suggest exposing the guidance interval as a hyperparameter in all diffusion models that use guidance. |
| title | Applying Guidance in a Limited Interval Improves Sample and Distribution Quality in Diffusion Models |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Neural and Evolutionary Computing |
| url | https://arxiv.org/abs/2404.07724 |