Matches in SemOpenAlex for { <https://semopenalex.org/work/W4307206816> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4307206816 abstract "Diffusion models can be viewed as mapping points in a high-dimensional latent space onto a low-dimensional learned manifold, typically an image manifold. The intermediate values between the latent space and image manifold can be interpreted as noisy images which are determined by the noise scheduling scheme employed during pre-training. We exploit this interpretation to introduce Boomerang, a local image manifold sampling approach using the dynamics of diffusion models. We call it Boomerang because we first add noise to an input image, moving it closer to the latent space, then bring it back to the image space through diffusion dynamics. We use this method to generate images which are similar, but nonidentical, to the original input images on the image manifold. We are able to set how close the generated image is to the original based on how much noise we add. Additionally, the generated images have a degree of stochasticity, allowing us to locally sample as many times as we want without repetition. We show three applications for which Boomerang can be used. First, we provide a framework for constructing privacy-preserving datasets having controllable degrees of anonymity. Second, we show how to use Boomerang for data augmentation while staying on the image manifold. Third, we introduce a framework for image super-resolution with 8x upsampling. Boomerang does not require any modification to the training of diffusion models and can be used with pretrained models on a single, inexpensive GPU." @default.
- W4307206816 created "2022-10-30" @default.
- W4307206816 creator A5008914922 @default.
- W4307206816 creator A5044553315 @default.
- W4307206816 creator A5068384228 @default.
- W4307206816 creator A5072713767 @default.
- W4307206816 creator A5075761244 @default.
- W4307206816 date "2022-10-21" @default.
- W4307206816 modified "2023-10-14" @default.
- W4307206816 title "Boomerang: Local sampling on image manifolds using diffusion models" @default.
- W4307206816 doi "https://doi.org/10.48550/arxiv.2210.12100" @default.
- W4307206816 hasPublicationYear "2022" @default.
- W4307206816 type Work @default.
- W4307206816 citedByCount "0" @default.
- W4307206816 crossrefType "posted-content" @default.
- W4307206816 hasAuthorship W4307206816A5008914922 @default.
- W4307206816 hasAuthorship W4307206816A5044553315 @default.
- W4307206816 hasAuthorship W4307206816A5068384228 @default.
- W4307206816 hasAuthorship W4307206816A5072713767 @default.
- W4307206816 hasAuthorship W4307206816A5075761244 @default.
- W4307206816 hasBestOaLocation W43072068161 @default.
- W4307206816 hasConcept C106131492 @default.
- W4307206816 hasConcept C110384440 @default.
- W4307206816 hasConcept C11413529 @default.
- W4307206816 hasConcept C115961682 @default.
- W4307206816 hasConcept C127413603 @default.
- W4307206816 hasConcept C13670688 @default.
- W4307206816 hasConcept C140779682 @default.
- W4307206816 hasConcept C151876577 @default.
- W4307206816 hasConcept C154945302 @default.
- W4307206816 hasConcept C31972630 @default.
- W4307206816 hasConcept C41008148 @default.
- W4307206816 hasConcept C529865628 @default.
- W4307206816 hasConcept C55128770 @default.
- W4307206816 hasConcept C70518039 @default.
- W4307206816 hasConcept C78519656 @default.
- W4307206816 hasConcept C99498987 @default.
- W4307206816 hasConceptScore W4307206816C106131492 @default.
- W4307206816 hasConceptScore W4307206816C110384440 @default.
- W4307206816 hasConceptScore W4307206816C11413529 @default.
- W4307206816 hasConceptScore W4307206816C115961682 @default.
- W4307206816 hasConceptScore W4307206816C127413603 @default.
- W4307206816 hasConceptScore W4307206816C13670688 @default.
- W4307206816 hasConceptScore W4307206816C140779682 @default.
- W4307206816 hasConceptScore W4307206816C151876577 @default.
- W4307206816 hasConceptScore W4307206816C154945302 @default.
- W4307206816 hasConceptScore W4307206816C31972630 @default.
- W4307206816 hasConceptScore W4307206816C41008148 @default.
- W4307206816 hasConceptScore W4307206816C529865628 @default.
- W4307206816 hasConceptScore W4307206816C55128770 @default.
- W4307206816 hasConceptScore W4307206816C70518039 @default.
- W4307206816 hasConceptScore W4307206816C78519656 @default.
- W4307206816 hasConceptScore W4307206816C99498987 @default.
- W4307206816 hasLocation W43072068161 @default.
- W4307206816 hasOpenAccess W4307206816 @default.
- W4307206816 hasPrimaryLocation W43072068161 @default.
- W4307206816 hasRelatedWork W1998618079 @default.
- W4307206816 hasRelatedWork W2005185696 @default.
- W4307206816 hasRelatedWork W2090346739 @default.
- W4307206816 hasRelatedWork W2161229648 @default.
- W4307206816 hasRelatedWork W2181461482 @default.
- W4307206816 hasRelatedWork W2235753890 @default.
- W4307206816 hasRelatedWork W2787555990 @default.
- W4307206816 hasRelatedWork W2990559186 @default.
- W4307206816 hasRelatedWork W2993674027 @default.
- W4307206816 hasRelatedWork W4307206816 @default.
- W4307206816 isParatext "false" @default.
- W4307206816 isRetracted "false" @default.
- W4307206816 workType "article" @default.