ScholarGate
Asistent
Machine learningDeep learning / NLP / CV

Slabo nadzirani difuzijski model

Slabo nadzirani difuzijski model trenira ili uvjetuje denojzirajući difuzijski vjerojatnosni model koristeći grube, bučne ili nepotpune nadzorne signale — poput oznaka klase na razini slike, graničnih okvira ili anotacija prikupljenih od mnoštva (crowd-sourced annotations) — umjesto pikselno precizne istine. To omogućuje visokokvalitetne generativne i diskriminativne izlaze u okruženjima s oskudnim anotacijama, gdje je potpuno označavanje neizvedivo ili pretjerano skupo.

Otvorite u MethodMindUskoroVideoUskoroDownload slides

Pročitajte cijelu metodu

Samo za članove

Prijavite se besplatnim računom kako biste pročitali ovaj odjeljak.

Prijavite se

Method map

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

Izvori

  1. Ho, J., Jain, A., & Abbeel, P. (2020). Denoising Diffusion Probabilistic Models. Advances in Neural Information Processing Systems (NeurIPS), 33, 6840–6851. link
  2. Zhou, K., et al. (2023). Weakly-supervised Semantic Segmentation with Diffusion Models. arXiv preprint arXiv:2309.11803. link

Kako citirati ovu stranicu

ScholarGate. (2026, June 3). Weakly Supervised Diffusion Model (Denoising Diffusion with Imperfect Supervision). ScholarGate. https://scholargate.app/hr/deep-learning/weakly-supervised-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
ScholarGateWeakly Supervised Diffusion Model (Weakly Supervised Diffusion Model (Denoising Diffusion with Imperfect Supervision)). Preuzeto 2026-06-15 s https://scholargate.app/hr/deep-learning/weakly-supervised-diffusion-model · Skup podataka: https://doi.org/10.5281/zenodo.20539026