Spandrel gives your project support for various PyTorch architectures meant for AI Super-Resolution, restoration, and inpainting. Based on the model support implemented in chaiNNer.
It would be interesting to see an implementation of ITER in Spandrel.
Iterative Token Evaluation and Refinement for Real-World Super-Resolutionhttps://github.com/chaofengc/ITER/tree/main
By the author of FemaSR architecture.
Personally, I've tried ITER in a conda environment, and it gives great results for some photos but not for others. Maybe I'll have more success in Chainner if you integrate ITER.
Unfortunately, the author temporarily deleted the x4 model (it was in the release section until last month :( ). Only x2 for now. Thanks.
It would be interesting to see an implementation of ITER in Spandrel. Iterative Token Evaluation and Refinement for Real-World Super-Resolution https://github.com/chaofengc/ITER/tree/main By the author of FemaSR architecture. Personally, I've tried ITER in a conda environment, and it gives great results for some photos but not for others. Maybe I'll have more success in Chainner if you integrate ITER. Unfortunately, the author temporarily deleted the x4 model (it was in the release section until last month :( ). Only x2 for now. Thanks.