Matches in SemOpenAlex for { <https://semopenalex.org/work/W3217030260> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W3217030260 abstract "Diffusion probabilistic models (DPMs) have achieved remarkable quality in image generation that rivals GANs'. But unlike GANs, DPMs use a set of latent variables that lack semantic meaning and cannot serve as a useful representation for other tasks. This paper explores the possibility of using DPMs for representation learning and seeks to extract a meaningful and decodable representation of an input image via autoencoding. Our key idea is to use a learnable encoder for discovering the high-level semantics, and a DPM as the decoder for modeling the remaining stochastic variations. Our method can encode any image into a two-part latent code where the first part is semantically meaningful and linear, and the second part captures stochastic details, allowing near-exact reconstruction. This capability enables challenging applications that currently foil GAN-based methods, such as attribute manipulation on real images. We also show that this two-level encoding improves denoising efficiency and naturally facilitates various downstream tasks including few-shot conditional sampling. Please visit our page: https://Diff-AE.github.io/" @default.
- W3217030260 created "2021-12-06" @default.
- W3217030260 creator A5000290961 @default.
- W3217030260 creator A5039154896 @default.
- W3217030260 creator A5065745608 @default.
- W3217030260 creator A5076638650 @default.
- W3217030260 date "2022-06-01" @default.
- W3217030260 modified "2023-10-10" @default.
- W3217030260 title "Diffusion Autoencoders: Toward a Meaningful and Decodable Representation" @default.
- W3217030260 cites W1834627138 @default.
- W3217030260 cites W2108598243 @default.
- W3217030260 cites W2133665775 @default.
- W3217030260 cites W2962785568 @default.
- W3217030260 cites W2963626105 @default.
- W3217030260 cites W3034431451 @default.
- W3217030260 cites W3159647175 @default.
- W3217030260 cites W3180196270 @default.
- W3217030260 cites W4226014430 @default.
- W3217030260 doi "https://doi.org/10.1109/cvpr52688.2022.01036" @default.
- W3217030260 hasPublicationYear "2022" @default.
- W3217030260 type Work @default.
- W3217030260 sameAs 3217030260 @default.
- W3217030260 citedByCount "22" @default.
- W3217030260 countsByYear W32170302602022 @default.
- W3217030260 countsByYear W32170302602023 @default.
- W3217030260 crossrefType "proceedings-article" @default.
- W3217030260 hasAuthorship W3217030260A5000290961 @default.
- W3217030260 hasAuthorship W3217030260A5039154896 @default.
- W3217030260 hasAuthorship W3217030260A5065745608 @default.
- W3217030260 hasAuthorship W3217030260A5076638650 @default.
- W3217030260 hasBestOaLocation W32170302602 @default.
- W3217030260 hasConcept C104317684 @default.
- W3217030260 hasConcept C111919701 @default.
- W3217030260 hasConcept C115961682 @default.
- W3217030260 hasConcept C118505674 @default.
- W3217030260 hasConcept C125411270 @default.
- W3217030260 hasConcept C153180895 @default.
- W3217030260 hasConcept C154945302 @default.
- W3217030260 hasConcept C177264268 @default.
- W3217030260 hasConcept C17744445 @default.
- W3217030260 hasConcept C184337299 @default.
- W3217030260 hasConcept C185592680 @default.
- W3217030260 hasConcept C199360897 @default.
- W3217030260 hasConcept C199539241 @default.
- W3217030260 hasConcept C204321447 @default.
- W3217030260 hasConcept C2776359362 @default.
- W3217030260 hasConcept C2776760102 @default.
- W3217030260 hasConcept C41008148 @default.
- W3217030260 hasConcept C49937458 @default.
- W3217030260 hasConcept C55493867 @default.
- W3217030260 hasConcept C66746571 @default.
- W3217030260 hasConcept C80444323 @default.
- W3217030260 hasConcept C94625758 @default.
- W3217030260 hasConceptScore W3217030260C104317684 @default.
- W3217030260 hasConceptScore W3217030260C111919701 @default.
- W3217030260 hasConceptScore W3217030260C115961682 @default.
- W3217030260 hasConceptScore W3217030260C118505674 @default.
- W3217030260 hasConceptScore W3217030260C125411270 @default.
- W3217030260 hasConceptScore W3217030260C153180895 @default.
- W3217030260 hasConceptScore W3217030260C154945302 @default.
- W3217030260 hasConceptScore W3217030260C177264268 @default.
- W3217030260 hasConceptScore W3217030260C17744445 @default.
- W3217030260 hasConceptScore W3217030260C184337299 @default.
- W3217030260 hasConceptScore W3217030260C185592680 @default.
- W3217030260 hasConceptScore W3217030260C199360897 @default.
- W3217030260 hasConceptScore W3217030260C199539241 @default.
- W3217030260 hasConceptScore W3217030260C204321447 @default.
- W3217030260 hasConceptScore W3217030260C2776359362 @default.
- W3217030260 hasConceptScore W3217030260C2776760102 @default.
- W3217030260 hasConceptScore W3217030260C41008148 @default.
- W3217030260 hasConceptScore W3217030260C49937458 @default.
- W3217030260 hasConceptScore W3217030260C55493867 @default.
- W3217030260 hasConceptScore W3217030260C66746571 @default.
- W3217030260 hasConceptScore W3217030260C80444323 @default.
- W3217030260 hasConceptScore W3217030260C94625758 @default.
- W3217030260 hasLocation W32170302601 @default.
- W3217030260 hasLocation W32170302602 @default.
- W3217030260 hasOpenAccess W3217030260 @default.
- W3217030260 hasPrimaryLocation W32170302601 @default.
- W3217030260 hasRelatedWork W140709781 @default.
- W3217030260 hasRelatedWork W176174999 @default.
- W3217030260 hasRelatedWork W2275988210 @default.
- W3217030260 hasRelatedWork W2359640100 @default.
- W3217030260 hasRelatedWork W2389073067 @default.
- W3217030260 hasRelatedWork W2760722755 @default.
- W3217030260 hasRelatedWork W3008321871 @default.
- W3217030260 hasRelatedWork W4320166805 @default.
- W3217030260 hasRelatedWork W4376652758 @default.
- W3217030260 hasRelatedWork W4378767251 @default.
- W3217030260 isParatext "false" @default.
- W3217030260 isRetracted "false" @default.
- W3217030260 magId "3217030260" @default.
- W3217030260 workType "article" @default.