Diffusion Probabilistic Modeling for Video Generation
Entropy · 20.10.2023 · 25 (10) · 1469
DOI: 10.3390/e25101469 ↗Kurz gefasst
Fachartikel in der Zeitschrift Entropy, online erschienen am 20.10.2023. Verfasst von Ruihan Yang und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, University of California, Irvine, CA 92697, USA. Die Arbeit wurde bisher 141-mal zitiert (Crossref) und nennt 94 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation (NSF), the Department of Energy, Office of Science, IARPA WRIVA program.
Abstract (Originalsprache)
Denoising diffusion probabilistic models are a promising new class of generative models that mark a milestone in high-quality image generation. This paper showcases their ability to sequentially generate video, surpassing prior methods in perceptual and probabilistic forecasting metrics. We propose an autoregressive, end-to-end optimized video diffusion model inspired by recent advances in neural video compression. The model successively generates future frames by correcting a deterministic next-frame prediction using a stochastic residual generated by an inverse diffusion process. We compare this approach against six baselines on four datasets involving natural and simulation-based videos. We find significant improvements in terms of perceptual quality and probabilistic frame forecasting ability for all datasets.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Computer Science, University of California, Irvine, CA 92697, USA.
ORCID-Profile: Ruihan Yang, Stephan Mandt
Förderung
- National Science Foundation (NSF)
- the Department of Energy, Office of Science
- IARPA WRIVA program
- Department of Energy, Office of Science
Zitieren
APA
Yang, R., Srivastava, P., & Mandt, S. (2023). Diffusion Probabilistic Modeling for Video Generation. Entropy, 25(10), 1469. https://doi.org/10.3390/e25101469
Vancouver
Yang R, Srivastava P, Mandt S. Diffusion Probabilistic Modeling for Video Generation. Entropy. 2023;25(10):1469. doi:10.3390/e25101469
BibTeX
@article{Yang2023,
author = {Yang, Ruihan and Srivastava, Prakhar and Mandt, Stephan},
title = {Diffusion Probabilistic Modeling for Video Generation},
journal = {Entropy},
year = {2023},
volume = {25},
number = {10},
pages = {1469},
doi = {10.3390/e25101469}
}
EUVIMED