A new LoRA (Low-Rank Adaptation) model called MOSS Sharpness LoRA has been developed for Stable Diffusion, aimed at controlling image sharpness and focus. The LoRA was trained on a dataset of 72 high-quality photographs, with some images intentionally blurred to teach the model the difference between in-focus and out-of-focus subjects. Users can adjust the LoRA's weight from 1 for sharp images to -1 for blurry images, offering control over depth of field. AI
IMPACT Provides users with finer control over image focus and sharpness in AI-generated art.
RANK_REASON This is a user-created add-on for an existing AI image generation model, not a release from a frontier lab.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
