ScholarGate
Assistent
Machine learning

Diffusionsmodel

En diffusionsmodel er en generativ deep-learning-metode, introduceret af Ho, Jain og Abbeel i 2020 (DDPM), der lærer at producere billeder, lyd og molekylære strukturer af høj kvalitet ved at vende en trinvis støjproces. Den har i vid udstrækning fortrængt GAN'er som den nuværende state-of-the-art inden for generativ modellering.

Åbn i MethodMindSnartVideoSnartDownload slides

Læs hele metoden

Kun for medlemmer

Log ind med en gratis konto for at læse dette afsnit.

Log ind

Method map

The neighbourhood of related methods — select a node to explore.

+1 more

Kilder

  1. Ho, J., Jain, A. & Abbeel, P. (2020). Denoising Diffusion Probabilistic Models. NeurIPS. link
  2. Rombach, R., Blattmann, A., Lorenz, D., Esser, P. & Ommer, B. (2022). High-Resolution Image Synthesis with Latent Diffusion Models. CVPR. link

Sådan citerer du denne side

ScholarGate. (2026, June 1). Denoising Diffusion Probabilistic Model (DDPM / Latent Diffusion). ScholarGate. https://scholargate.app/da/deep-learning/diffusion-model

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

Compare side by side

Refereret af

ScholarGateDiffusion Model (Denoising Diffusion Probabilistic Model (DDPM / Latent Diffusion)). Hentet 2026-06-15 fra https://scholargate.app/da/deep-learning/diffusion-model · Datasæt: https://doi.org/10.5281/zenodo.20539026