GIMP-ML VS maxim

Uh an MTG player here, also a freak of GIMP. Also a Computer Vision Engineer, up to my knowledge there exists a ML framework plugin for GIMP 2 called GIMP ML, which works on both Mac and Linux: https://github.com/kritiksoman/GIMP-ML, but I really do not know the internals of the plugin and I do not know if there is an interface to do this.

Just like Photoshop, GIMP is only capable of fast and complex editing with the help of brushes(here are some good ones), scripts and sometimes plugins(also this one is recommended, available on most distro's repos as both standalone or as a plugin).

For RAW editing i could suggest RawTherapee as an extension to GIMP, amazing tool and you'll feel home. GIMP supports python extensions and there are actually a bunch of AI supported extensions for GIMP, one project called GIMP-ML actually collects AI extensions together for GIMP, and they're more safer than Photoshop since they're not sending images to Adobe servers, all processed locally and depending on your computer faster than Photoshop(Since GIMP doesn't wait response from outside servers).

The best place to report this would be the GIMP-ML issue tracker at https://github.com/kritiksoman/GIMP-ML/issues

Github: https://github.com/kritiksoman/GIMP-ML/tree/GIMP3-ML

For example, the official instructions at https://github.com/kritiksoman/GIMP-ML/blob/master/INSTALLATION.md are certainly not written for the absolute beginner who will likely not be familiar with anything mentioned there, but someone who knows the basics of installing Python packages with pip and virtualenv will start to follow them and may run into problems along the way.

Today we present a new multi-axis approach that is simple and effective, improves on the original ViT and MLP models, can better adapt to high-resolution, dense prediction tasks, and can naturally adapt to different input sizes with high flexibility and low complexity. Based on this approach, we have built two backbone models for high-level and low-level vision tasks. We describe the first in “MaxViT: Multi-Axis Vision Transformer”, to be presented in ECCV 2022, and show it significantly improves the state of the art for high-level tasks, such as image classification, object detection, segmentation, quality assessment, and generation. The second, presented in “MAXIM: Multi-Axis MLP for Image Processing” at CVPR 2022, is based on a UNet-like architecture and achieves competitive performance on low-level imaging tasks including denoising, deblurring, dehazing, deraining, and low-light enhancement. To facilitate further research on efficient Transformer and MLP models, we have open-sourced the code and models for both MaxViT and MAXIM.