Janspiry/Image-Super-Resolution-via-Iterative-Refinement
PyTorch implementation of SR3 (Image Super-Resolution via Iterative Refinement), a diffusion probabilistic model for upscaling images from low to high resolution.
Not currently ranked — collecting fresh signals.
star history
This repository provides an unofficial implementation of the SR3 paper, using diffusion probabilistic models (DDPM) for single-image super-resolution. The model can upscale images such as 64x64 to 512x512 or 16x16 to 128x128 using a learned reverse diffusion process. It includes training pipelines with multi-GPU support, evaluation metrics, and pretrained model checkpoints.
Frequently asked
- What is Janspiry/Image-Super-Resolution-via-Iterative-Refinement?
- PyTorch implementation of SR3 (Image Super-Resolution via Iterative Refinement), a diffusion probabilistic model for upscaling images from low to high resolution.
- Is Image-Super-Resolution-via-Iterative-Refinement open source?
- Yes — Janspiry/Image-Super-Resolution-via-Iterative-Refinement is open source, released under the Apache-2.0 license.
- What language is Image-Super-Resolution-via-Iterative-Refinement written in?
- Janspiry/Image-Super-Resolution-via-Iterative-Refinement is primarily written in Python.
- How popular is Image-Super-Resolution-via-Iterative-Refinement?
- Janspiry/Image-Super-Resolution-via-Iterative-Refinement has 3.9k stars on GitHub.
- Where can I find Image-Super-Resolution-via-Iterative-Refinement?
- Janspiry/Image-Super-Resolution-via-Iterative-Refinement is on GitHub at https://github.com/Janspiry/Image-Super-Resolution-via-Iterative-Refinement.