CleanDIFT: Diffusion Features without Noise

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Stracke, Nick, Baumann, Stefan Andreas, Bauer, Kolja, Fundel, Frank, Ommer, Björn
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909960973058048
author Stracke, Nick
Baumann, Stefan Andreas
Bauer, Kolja
Fundel, Frank
Ommer, Björn
author_facet Stracke, Nick
Baumann, Stefan Andreas
Bauer, Kolja
Fundel, Frank
Ommer, Björn
contents Internal features from large-scale pre-trained diffusion models have recently been established as powerful semantic descriptors for a wide range of downstream tasks. Works that use these features generally need to add noise to images before passing them through the model to obtain the semantic features, as the models do not offer the most useful features when given images with little to no noise. We show that this noise has a critical impact on the usefulness of these features that cannot be remedied by ensembling with different random noises. We address this issue by introducing a lightweight, unsupervised fine-tuning method that enables diffusion backbones to provide high-quality, noise-free semantic features. We show that these features readily outperform previous diffusion features by a wide margin in a wide variety of extraction setups and downstream tasks, offering better performance than even ensemble-based methods at a fraction of the cost.
format Preprint
id arxiv_https___arxiv_org_abs_2412_03439
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CleanDIFT: Diffusion Features without Noise
Stracke, Nick
Baumann, Stefan Andreas
Bauer, Kolja
Fundel, Frank
Ommer, Björn
Computer Vision and Pattern Recognition
Internal features from large-scale pre-trained diffusion models have recently been established as powerful semantic descriptors for a wide range of downstream tasks. Works that use these features generally need to add noise to images before passing them through the model to obtain the semantic features, as the models do not offer the most useful features when given images with little to no noise. We show that this noise has a critical impact on the usefulness of these features that cannot be remedied by ensembling with different random noises. We address this issue by introducing a lightweight, unsupervised fine-tuning method that enables diffusion backbones to provide high-quality, noise-free semantic features. We show that these features readily outperform previous diffusion features by a wide margin in a wide variety of extraction setups and downstream tasks, offering better performance than even ensemble-based methods at a fraction of the cost.
title CleanDIFT: Diffusion Features without Noise
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.03439