Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386066475> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W4386066475 abstract "Despite the remarkable progress in deep generative models, synthesizing high-resolution and temporally coherent videos still remains a challenge due to their high-dimensionality and complex temporal dynamics along with large spatial variations. Recent works on diffusion models have shown their potential to solve this challenge, yet they suffer from severe computation and memory-inefficiency that limit the scalability. To handle this issue, we propose a novel generative model for videos, coined projected latent video diffusion model (PVDM), a probabilistic diffusion model which learns a video distribution in a low-dimensional latent space and thus can be efficiently trained with high-resolution videos under limited resources. Specifically, PVDM is composed of two components: (a) an autoencoder that projects a given video as 2D-shaped latent vectors that factorize the complex cubic structure of video pixels and (b) a diffusion model architecture specialized for our new factorized latent space and the training/sampling procedure to synthesize videos of arbitrary length with a single model. Experiments on popular video generation datasets demonstrate the superiority of PVDM compared with previous video synthesis methods; e.g., PVDM obtains the FVD score of 639.7 on the UCF-101 long video (128 frames) generation benchmark, which improves 1773.4 of the prior state-of-the-art." @default.
- W4386066475 created "2023-08-23" @default.
- W4386066475 creator A5015735286 @default.
- W4386066475 creator A5035397475 @default.
- W4386066475 creator A5044591479 @default.
- W4386066475 creator A5072573916 @default.
- W4386066475 date "2023-06-01" @default.
- W4386066475 modified "2023-10-11" @default.
- W4386066475 title "Video Probabilistic Diffusion Models in Projected Latent Space" @default.
- W4386066475 cites W1522734439 @default.
- W4386066475 cites W2016053056 @default.
- W4386066475 cites W2962785568 @default.
- W4386066475 cites W2963092440 @default.
- W4386066475 cites W2964245526 @default.
- W4386066475 cites W2994586138 @default.
- W4386066475 cites W2994903658 @default.
- W4386066475 cites W3031246127 @default.
- W4386066475 cites W3035574324 @default.
- W4386066475 cites W3120341153 @default.
- W4386066475 cites W3180196270 @default.
- W4386066475 cites W3180355996 @default.
- W4386066475 cites W4308233907 @default.
- W4386066475 cites W4312423208 @default.
- W4386066475 cites W4312933868 @default.
- W4386066475 cites W4382203549 @default.
- W4386066475 doi "https://doi.org/10.1109/cvpr52729.2023.01770" @default.
- W4386066475 hasPublicationYear "2023" @default.
- W4386066475 type Work @default.
- W4386066475 citedByCount "1" @default.
- W4386066475 crossrefType "proceedings-article" @default.
- W4386066475 hasAuthorship W4386066475A5015735286 @default.
- W4386066475 hasAuthorship W4386066475A5035397475 @default.
- W4386066475 hasAuthorship W4386066475A5044591479 @default.
- W4386066475 hasAuthorship W4386066475A5072573916 @default.
- W4386066475 hasConcept C101738243 @default.
- W4386066475 hasConcept C108583219 @default.
- W4386066475 hasConcept C111030470 @default.
- W4386066475 hasConcept C112933361 @default.
- W4386066475 hasConcept C13280743 @default.
- W4386066475 hasConcept C153180895 @default.
- W4386066475 hasConcept C154945302 @default.
- W4386066475 hasConcept C160633673 @default.
- W4386066475 hasConcept C167966045 @default.
- W4386066475 hasConcept C185798385 @default.
- W4386066475 hasConcept C205649164 @default.
- W4386066475 hasConcept C31972630 @default.
- W4386066475 hasConcept C39890363 @default.
- W4386066475 hasConcept C41008148 @default.
- W4386066475 hasConcept C48044578 @default.
- W4386066475 hasConcept C49937458 @default.
- W4386066475 hasConcept C77088390 @default.
- W4386066475 hasConceptScore W4386066475C101738243 @default.
- W4386066475 hasConceptScore W4386066475C108583219 @default.
- W4386066475 hasConceptScore W4386066475C111030470 @default.
- W4386066475 hasConceptScore W4386066475C112933361 @default.
- W4386066475 hasConceptScore W4386066475C13280743 @default.
- W4386066475 hasConceptScore W4386066475C153180895 @default.
- W4386066475 hasConceptScore W4386066475C154945302 @default.
- W4386066475 hasConceptScore W4386066475C160633673 @default.
- W4386066475 hasConceptScore W4386066475C167966045 @default.
- W4386066475 hasConceptScore W4386066475C185798385 @default.
- W4386066475 hasConceptScore W4386066475C205649164 @default.
- W4386066475 hasConceptScore W4386066475C31972630 @default.
- W4386066475 hasConceptScore W4386066475C39890363 @default.
- W4386066475 hasConceptScore W4386066475C41008148 @default.
- W4386066475 hasConceptScore W4386066475C48044578 @default.
- W4386066475 hasConceptScore W4386066475C49937458 @default.
- W4386066475 hasConceptScore W4386066475C77088390 @default.
- W4386066475 hasFunder F4320324161 @default.
- W4386066475 hasFunder F4320335489 @default.
- W4386066475 hasLocation W43860664751 @default.
- W4386066475 hasOpenAccess W4386066475 @default.
- W4386066475 hasPrimaryLocation W43860664751 @default.
- W4386066475 hasRelatedWork W112744582 @default.
- W4386066475 hasRelatedWork W121273120 @default.
- W4386066475 hasRelatedWork W2292254049 @default.
- W4386066475 hasRelatedWork W2337415362 @default.
- W4386066475 hasRelatedWork W2740820121 @default.
- W4386066475 hasRelatedWork W2897995864 @default.
- W4386066475 hasRelatedWork W2998168123 @default.
- W4386066475 hasRelatedWork W317572212 @default.
- W4386066475 hasRelatedWork W4287995534 @default.
- W4386066475 hasRelatedWork W4312857205 @default.
- W4386066475 isParatext "false" @default.
- W4386066475 isRetracted "false" @default.
- W4386066475 workType "article" @default.