SUPIR: New SOTA Open Source Image Upscaler & Enhancer Model Better Than Magnific & Topaz AI Tutorial

With V8, NOW WORKS on 12 GB GPUs as well with Juggernaut-XL-v9 base model. In this tutorial video, I introduce SUPIR (Scaling-UP Image Restoration), a state-of-the-art image enhancing and upscaling model presented in the paper "Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild." SUPIR surpasses the performance of expensive alternatives like Magnific AI or Topaz AI and is open-source, with the models readily available. Additionally, I provide a one-click installer for easy installation and use on various platforms, including Windows, RunPod, and Linux. SUPIR also incorporates the Stable Diffusion XL (SDXL) pipeline for superior photo upscaling and enhancement.

#SUPIR #StableDiffusion #SDXL

The Patreon Post Link Used In The Video To Download Installers ⤵️

Official GitHub Link ⤵️

Our Discord Channel ⤵️

Our Patreon With Amazing AI Scripts & Tutorials ⤵️

0:00 Introduction to SUPIR (Scaling-UP Image Restoration) full tutorial
2:10 How to download and install SUPIR on Windows or RunPod (thus Linux)
3:19 How to setup a community Pod on RunPod's newest interface
4:33 How to install and start SUPIR on RunPod
7:10 How to use Proxy connect of RunPod
8:13 How to install and start our own quantization supporting LLaVA
9:22 Getting image description from our own LLaVA model
9:42 How to use SUPIR interface and testing camel image (test image 1) on SUPIR in details
12:07 Testing a very old family photo enhancement and upscaling with SUPIR (test image 2)
14:34 Where the generated images are saved
14:53 Testing the image of Arnold Schwarzenegger as a warrior (test image 3) on SUPIR in details
16:22 The effect of simple prompt vs detailed prompt
17:30 Testing a dragon statue enhancement and upscaling with SUPIR (test image 4)
17:42 How I used ChatGPT Plus / GPT-4 for image captioning
18:29 The model works with literally every resolution and example very big upscale
19:00 Testing image of a dinosaur in jurassic park image enhancement and upscaling with SUPIR (test image 5)
19:41 From 500px to 3000px upscale results and how to do very big upscale properly
22:39 GPU utilization of the SUPIR scripts
23:15 If you get out of VRAM error what can you do and how you can solve
25:22 Testing a MonsterMMORPG Game character (anime like drawing) upscaling and image enhancing (test image 6)
25:39 What to do if your image has transparent pixels to be able to upscale
27:35 Testing a black and white colored movie screenshot of a man image enhancement and upscaling with SUPIR (test image 7)
28:29 Testing a screenshot from the movie Predator enhancement and upscaling with SUPIR (test image 8)
29:12 The queue ability of the Gradio app of SUPIR
29:49 Testing an old photo of Muhammad Ali in a boxing stance image enhancement and upscaling with SUPIR (test image 9)
30:45 Testing a black and white colored movie screenshot of Charlie Chaplin image enhancement and upscaling with SUPIR (test image 10)

Info From The Paper

Sure, here's a summary of the paper "Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild" (SUPIR), with the goal of at least 3,000 characters.

The paper introduces SUPIR, a groundbreaking image restoration (IR) approach that combines a powerful generative prior with the benefits of model scaling. SUPIR leverages multi-modal techniques and a large-scale generative prior, making significant strides towards intelligent and realistic image restoration. The authors demonstrate SUPIR's superiority in various IR tasks, achieving exceptional visual quality. A key innovation is the model scaling technique, offering dramatic improvements in capabilities and pushing the boundaries of image restoration. Additionally, the model offers the unique ability to be controlled via text prompts, greatly expanding its applications and potential.

Advanced Generative Prior: SUPIR utilizes StableDiffusion-XL (SDXL), a massive generative model with 2.6 billion parameters. SDXL serves as a powerful tool for introducing high-quality image generation abilities into the image restoration process.

Image Encoder Fine-Tuning: The image encoder is fine-tuned to improve its resilience to image degradations, ensuring robust interpretation of low-quality input images.

Large-Scale Training Dataset: A massive dataset comprising 20 million high-resolution, high-quality images is collected to fully harness the potential of model scaling. Descriptive text annotations accompany each image, enabling text-based control of image restoration.

Multi-modal Language Integration: A 13-billion-parameter multi-modal language model is used to provide descriptive prompts of image content, greatly enhancing the model's ability to understand and restore images accurately.

[SECourses]

With V8, NOW WORKS on 12 GB GPUs as well with Juggernaut-XL-v9 base model

[tdfilmstudio]

I think you finally reached the next level of brilliance with this! Something superior to the best competitor tools out there! Congratulations and a huge thanks

[HealthNutriNexus]

No one else is talking about this tool on youtube. Thank you very much! 😀👍

[neogeo]

I have to say, you have done such a great job updating your branch to add features to this. It has now become indispensable in my dataset prep workflow. I can't recommend your work enough.

[XavierCliment]

certainly is a next level!! The original SUPIR application needs at least 32 GB of VRAM, Dr.Furkan reduce with 12 GB VRAM, and it's constantly working to improve, as a patreon member i'm very satisfied with his works

[paidoclock]

Wow this is too amazing!! Thank you so much for installer

[dreamzdziner]

You sir are doing amazing work by doing such tutorials to the AI universe freely. Respect!🧡🙏

[iresolvers]

do you have a comfyui version of this workflow??...great video!

[brianmonarchcomedy]

Just to be clear... We need to join patreon to get the easier installer? Thanks

Hocam ne arasam karşıma siz çıkıyorsunuz. Dünyada bu konuda en iyilerden birisiniz. Benim merak ettiğim bir şey var. Öncelikle bu upscale ve kalite artırma işlemleri görüntünün bir kısmının istemeden de olsa bozulmasını sağlıyor. Burada karakter tutarlılığı sağlanabilseydi video restorasyonunda (doğru sonuç için çok fazla parametre var ve bu kadar zaman ve kaynağa erişimimiz şimdlik yok.) çok iyi sonuçlar çıkarılabilirdi. Sorum şu; acaba bildiğiniz video restorasyonu ile ilgili yapılan bir ai çalışması var mı? Özellikle de Supir'e benzer bir altyapısı olan. Ya da Supir'i fotoğrafları sekans şeklinde verip çalıştırmak mümkün mü? Yanıtınız için Teşekkür ederim.

[smartkhawar]

Great lecture. Always interesting stuff. Brother, you are great.

[BuckeyeGuy]

This is a stunningly powerful tool. But 48GB VRAM just seems insanely out of reach. If someone can get this down to 12GB, it will be an absolute miracle.

[cleverestx]

Amazing results, except for the artificial wrinkle lines it adds underneath eyes...I can't unsee those now.

[ultimategolfarchives]

I was really intrigued with this, and Confyui still can do better with much lower vram and more control. But you need more patience haha 😂

Great breakdown

[pierruno]

I use it with Pinokio and it does not work

[rbdesignguy]

Man this is an awesome feature. Just wish they didn’t make you have to sign an nda to generate for your clients photos..

[cemilhaci]

Hocam eğitim için tşekkür ederim, patreondan üyelik almayı palnlıyorum, öncesinde kafama takılan bir kaç şeyi sormak isterim, patreon üyeliği ile indirdiğimiz dosyalar ile localde sizin yaptığımız işlemleri yapabilicez değilmi herhangi bir kısıtlama olmadan ve anladığım kadarı ile automatic1111 üzerinde değilde standalone tarzında çalıyor doğrumudur. Şimdiden teşekkür ederim

[ThoughtFission]

Will your one click install work if I already have other interfaces installed (with python) like automatic 1111, comfyui and Fooocus? I've had problems with one click installer before because they always want to overlay thier own python. Is there a Venv?

[ZAQeN]

Will it be perfect with Midjourney generated images ?

[ozama]

So, no free way to know how to install a free software on Windows?