GitHub - YingqingHe/ScaleCrafter: Official implementation of...

GitHub - YingqingHe/ScaleCrafter: Official implementation of ScaleCrafter for higher-resolution visual generation at inference time.

YingqingHe github.com

RelatedInsightsHighlights

So what tricks do we have for making stable diffusion better? - Multi-aspect-ratio training on nice images - Offset noise - ControlNet for extra control - Attend and excite - Guidance (CLIP, aesthetic, SAG, ?) - Upscaler What am I missing? Who has all... See more

Jonathan Whitaker x.com

eDiffi: Text-to-Image Diffusion Models with an Ensemble of Expert Denoisers abs: https://t.co/QaJfNzXeLL project page: https://t.co/UrYmzGek8C sota t2i diffusion model that consists of a base diffusion model and two super-resolution modules, 1024 × 1024 high-definition outputs... See more

AK x.com

Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding project page: https://t.co/6nzZPACkzV sota FID(7.27 on COCO), without ever training on COCO, human raters find Imagen samples to be on par with the COCO data itself in image-text alignment https://t.co/nIGyMwQLnX

x.com