Welcome to the exciting world of image processing! In this guide, we’ll explore how to use SupPixel AI, based on the innovative SUPIR technology, to upscale images and restore them with incredible photo-realistic quality. Whether you’re a photographer seeking to enhance your work or a digital artist looking to breathe life into your images, this article is for you!

Installation and Dependencies

To start using SupPixel AI, you’ll need to install some dependencies and set up your environment. Think of this as preparing your kitchen before you start cooking a delicious meal. Follow these steps:

1. Clone the Repository:

   git clone https://github.com/Fanghua-Yu/SUPIR.git
   cd SUPIR

2. Create a New Environment:

   conda create -n SUPIR python=3.8 -y
   conda activate SUPIR
   pip install --upgrade pip
   pip install -r requirements.txt

3. Download Checkpoints:

   If you can connect to Hugging Face, set the following paths in CKPT_PTH.py as None. This will allow the necessary models to be downloaded automatically!

Setting Up Dependent Models

Much like gathering all the ingredients for your favorite dish, you’ll need some models. Here’s a list to help you out:

• SDXL CLIP Encoder-1
• SDXL CLIP Encoder-2
• SDXL base 1.0_0.9vae
• LLaVA CLIP
• LLaVA v1.5 13B

Quick Inference and Usage

Now that everything is set up, it’s time to put the power of SupPixel AI to work. Imagine you are an artist, transforming a blank canvas into a masterpiece. Here’s how you can invoke the magic:

Basic Usage Command

python test.py [options]

Include various options for your upscaling process, such as:

• --img_dir – Specify the input folder containing your images.
• --save_dir – Designate where to save the processed images.
• --upscale – Set the upscaling ratio (default is 1).

Example Commands

• For best quality:

  CUDA_VISIBLE_DEVICES=0,1 python test.py --img_dir your_input_folder --save_dir .results-Q --upscale 2 --SUPIR_sign Q

• For light degradation:

  CUDA_VISIBLE_DEVICES=0,1 python test.py --img_dir your_input_folder --save_dir .results-F --SUPIR_sign F --upscale 2 --s_cfg 4.0 --linear_CFG

Troubleshooting Tips

If you encounter any issues, don’t fret! Here are some common troubleshooting steps:

• Ensure Python and Conda are properly installed. If not, re-install them.
• Check your internet connection if models are not downloading.
• Make sure you’ve set the paths correctly in the configuration files.

For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.

Conclusion

At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations. Dive into the realm of SupPixel AI, and elevate your image processing journey!