How to Upscale Your AI Art

In this newsletter, read about:

• 🕵️‍♀️ Best Tools for Image Upscaling

• 🗞 News and Top Reads

• 📌 AI Art Tutorial: AI Photography Guide

• 🎨 Featured Artist: Khaled Sadeden

• 🖼 AI-Assisted Artwork of the Week

• 🤓 How to Get Started with AI Art?

🕵️‍♀️ Best Tools for Image Upscaling

Artificial Intelligence has radically transformed the digital arts domain, with tools like Midjourney and Stable Diffusion creating incredibly realistic images. These AI tools have brought a remarkable revolution, yet there is a caveat; the quality of the generated images is still too low for printing and sale. For instance, the upscaled images from Midjourney are approximately 1024×1024 pixels if squared or 960 x 1200 pixels with a 4:5 aspect ratio - which is not sufficient for printing on a standard 16x20 size. Therefore, if you're an artist looking to print your AI art, upscaling the images using external tools becomes essential. Thankfully, we have numerous top-notch AI-powered image upscalers today.

Let's delve into some of the best and test them with upscaling the following image, generated with Midjourney.

Top 5 Image Upscalers for AI Art

1. Photo AI by Topaz Labs

Topaz Photo AI (an upgraded version of Gigapixel AI) is one of the most esteemed image upscalers in the AI art community. It has earned excellent reviews for the superior quality of its upscaled images. However, it lacks a browser interface, meaning that it requires downloading the software, which could take up a significant amount of space – at least, a few gigabytes. Moreover, it demands a fairly powerful GPU and requires a one-time payment of $199. The software also offers a trial mode, but the final images will be saved with a watermark. Despite these drawbacks, the remarkable quality of the upscaled images makes Topaz Photo AI a worthy consideration.

2. Upscayl

Upscayl is a free and open-source AI Image Upscaler. It requires a download but takes about four times less space than Topaz Photo AI. Upscayl is quite user-friendly, offering the option to choose between different models depending on which one works better for a specific image. The quality of the upscaled images might not match up to Topaz’s product, but considering that it's free, Upscayl is a strong contender in the image upscaler market.

3. Bigjpg

Bigjpg is an image enlarger that can be used via a browser-based interface or downloaded as an app. It offers a free version, but for faster and more stable enlarging, you need to become a paid user. The paid plans start from $6/month, including all the premium features for up to 500 images per month.

When upscaling, users can choose the type of artwork and the level of noise removal. This feature is particularly useful for AI-generated paintings where brush strokes need to remain visible on the upscaled image. While Bigjpg may not compete with the state-of-the-art upscalers, it performs exceptionally well with anime images and illustrations.

4. Icons8

Icons8 is a powerful AI image upscaler that offers limited free use with watermarked images. However, it does have flexible pricing, such as $0.20 per image or $9 per month for up to 100 images, or a one-time fee of $99. Icons8 has a browser interface but with limited functionality, and for the full experience, downloading an app is necessary. The tool can upscale images up to 7680x7680px, and the quality of upscaled images, although variable depending on the style of the image, can be excellent with minimal quality loss.

5. Waifu2x

Waifu2x is a free, open-source upscaling technology that uses Nvidia CUDA. Initially developed as an anime picture upscaler, it performs best with these types of images. Its performance with realistic photos is not as good. With Waifu2x, you can upscale images up to 10x and also set the level of noise reduction to suit your specific images. Given that it’s free, Waifu2x is a good tool to consider, but probably only if you are working with anime images and illustrations.

Performance Comparison

For this overview, I have been using only the free versions of the upscalers. Where the settings allowed, I was choosing 4x upscale with no noise reduction. The free version of Icons8 allows only 2x upscale, so it’s comparison to other tools was not very fair, but I’ve indicated this in the caption.

So which tool is your favorite? I love Topaz Photot AI and consider now buying full access to get rid of a watermark 😃 Upscayl also looks good, but seems to be a little bit less sharp than Topaz’s upscaler. I was also pleasantly surprised with the Bigjps’s performance – it doesn’t preserve as many details as Topaz Photo AI but makes the patterns look nice and sharp. Waifu2x is apparently not suitable for this kind of image.

To Sum Up

AI-powered image upscalers have made a huge impact on how we can manipulate and improve digital images. They offer a variety of options to AI creators, who can choose the best tool based on their specific needs. While some offer more sophisticated features and superior quality at a cost, others provide reasonable quality for free or at a lesser price. Remember that the goal is not just to increase the size of the image, but to do so in a way that maintains, or even improves, its quality. With state-of-the-art AI image upscalers, your images will be ready for high-quality prints or sale via the top image stocks websites.

🗞 News and Top Reads

• Bard chatbot, which is Google’s response to ChatGPT and Bing, will become much more visual in the coming months.

  • Adobe Firefly will be integrated into Bard, enabling text-to-image generation through chatting with Bard.

  • When applicable, Bard’s responses will include rich visuals to give you a much better sense of what you’re exploring.

  • Thanks to Google Lens integration, users will be also able to include images — alongside text — in their prompts.

• Google also wants to make spotting AI-generated images easier.

  • First of all, all the images generated using Google's own tools (e.g., Bard) will be clearly marked as AI-generated.

  • Even more interesting, the company says it has buy-in from Midjourney and Shutterstock on this, with others set to follow.

  • Finally, coming in the next few months and initially rolling out in the US, the tool called “About this image” will help people identify the original source of each image in Google Image Search and where else it’s been published. Spreading AI-altered fakes and AI-generated images as real ones will become much more challenging.

• Stability AI has released a text-to-animation model.

  • The company behind Stable Diffusion has launched another tool that lets people create animations using text, image, and video prompts.

  • There are three ways to create animations: (1) text to animation, (2) text input + initial image input, and (3) input video + text input.

  • Check out a demo video here.

📌 AI Art Tutorial: AI Photography Guide

In this 10-minute tutorial, Cihan offers a comprehensive guide on creating stunning, ultra-realistic images using Midjourney V5.1. The video covers everything from writing effective prompts with ChatGPT, leveraging the 'describe' function and multi-prompting techniques, exploring technical aspects and visual styles of AI photography, and then, upscaling the final images with AI-powered upscalers.

🎨 Featured Artist: Khaled Sadeden

Khaled Sadeden is an architectural designer based in Doha, Qatar. He has over 12 years of experience in architecture and urban design with a focus on computational design. Recently, Khaled decided to combine his exceptional expertise in architecture with the generative capabilities of AI to create stunning AI-generated designs. See more examples on Instagram @khaled_sadeden_design.

If you want to recommend an artist to be featured in this newsletter, feel free to respond to this email with the links to the artist’s website or Instagram profile. Self-promotion is also allowed 😃 

🖼 AI-Assisted Artwork of the Week

🤓 How to Get Started with AI Art?

1. DALL-E: Creating Images from Text – introduction to text-to-image generation.

2. The DALL-E 2 Prompt Book – a guidebook by OpenAI that explains how to effectively right prompts to generate images across different domains (e.g., photography, illustration, art history, 3D artwork).

3. All-In-One Guide For Midjourney: The Art Of Prompts – a comprehensive guide on creating effective Midjourney prompts by Karol Andruszków.

4. Best Midjourney Prompts – a guide that covers the basics of Midjourney prompts (e.g., which keywords to use to create abstract art, surreal art, minimalism, etc) as well as some more advanced options (e.g., keywords related to camera lenses and filters, imitating certain artists and photographers without using their names). Finally, they provide a list of 600+ creative text prompts for image generation.

5. Stable Diffusion Prompt Book – a prompt book prepared by OpenArt. The book discusses ideal prompt format, using modifiers to change the style, format, or perspective of the image, applying ”magic words” to improve image quality, adding negative prompts, and adjusting Stable Diffusion parameters.

Share Kiki and Mozart

If you like this newsletter and know somebody who might also like it, feel free to share this newsletter. Let’s have more people learn about AI art!

If you have been forwarded this email and you like it, please subscribe below. And welcome to the world of AI art!