News for hackers. Stuff that matters.
Re:Direct pixel-space megapixel image generation with diffus... by Birch-san on Tuesday January 23, 2024 @07:28PM (#39111994)
FID doesn't reward high-resolution detail. the inception feature size is 299x299! so we are forced to downsample our FFHQ-1024 samples to compute FID.

it also doesn't punish poor detail either! this advantages latent diffusion, which can claim to achieve a high resolution but without actually needing to have correct textures to get good metrics.

There are no comments on this comment yet.