Given a corrupted input image, Con\textit{ffusion}, repurposes a pretrained diffusion model to generate lower and upper bounds around each reconstructed pixel. The true pixel value is guaranteed to fall within these bounds with probability $p$.
Source: Conffusion: Confidence Intervals for Diffusion ModelsPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Conformal Prediction | 1 | 16.67% |
Facial Inpainting | 1 | 16.67% |
Image Generation | 1 | 16.67% |
Image Inpainting | 1 | 16.67% |
Prediction Intervals | 1 | 16.67% |
Super-Resolution | 1 | 16.67% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |