search

levihsu / OOTDiffusion

Official implementation of OOTDiffusion: Outfitting Fusion based Latent Diffusion for Controllable Virtual Try-on
Other
5.07k stars 740 forks source link

Generate single best image #161

Closed mirfan899 closed 2 months ago

mirfan899 commented 2 months ago

Currently, model gives 3 images as output and takes time in generation. Can we reduce the images to single best image???

levihsu commented 2 months ago

Hi. You can alter the sample number (1~6 for 24GB GPU) as you wish