Articles on: Image Enhancement

Which operations work best for different types of photos?

Our networks are built to serve different purposes: from UGC, landscapes, product photos to digital art. Choose the best operation to get optimal results.

💡 To get a high quality image, it's important to make sure it hasn't been previously upscaled with any tool.

If your image has already been upscaled or scanned from the physical original, sometimes reducing the resolution before processing can yield much better results.

The core basics about Let's Enhance algorithms

Let's Enhance algorithms help improve your photos in many ways: apply JPEG noise reduction (removing pixelation in your compressed images), bring out shapes, make your images crisp and sharp, fill in missing details, you name it. However, to get the best results, it is worth knowing which operations to choose for relevant photo types: taken by phone, digital camera, small compressed & pixelated, for printing, photos of people, architecture, cartoons, etc.

Choose the right algorithm that is best for your use case

OPERATIONS

Within the OPERATIONS tab, you can adjust the Upscale type, Remove background, apply Light AI (HDR effect), enhance Tone and Color. These can be applied accordingly within the menu panel on the right:

1. Gentle

The Gentle algorithm is trained to enhance images with all types of text, such as labels on product packaging, posters, and maps

2. Balanced

Balanced style specializes in low quality images (compressed, blurred, pixelated, etc.). It works wonders on low-quality product images and images with faces, but this neural network cannot recognize images with bodies. It is better to use the Strong or Gentle.

Balanced style works best on:

small & bad quality product photos
real estate
food for the web

💡 The use of the Balanced algorithm is not recommended for good and high quality images because this style is trained to aggressively restore missing details and intensively redraw them in low quality images. That's why some artifacts can be found in good/high quality images due to the over-intensive restoration.

3. Strong

low quality, pixelated images to up to 2.5MP
under development, so results may vary
Enhance Faces for face details restoration

4. Magic

Magic is our most powerful, next-generation AI upscaler for AI art, food and landscape photos

💡 Magic network is still in beta phase. Read more about this upscaler here: What is the Magic next-gen AI upscaler and how to use it?

5. Digital Art

Upscaling type Digital Art works best on:

drawings
illustrations
paintings
cartoons
anime

Use Digital Art for low-resolution illustrations and digital art

There are also other functions within the Advanced options tap you can adjust accordingly to your requirements:

Remove background

Use it to remove the background from your product, cars or general images. Remember to set the output file to PNG before processing to download the image with the transparent background. Otherwise it will be converted to JPEG and the background will be white.

Light AI

Use the Light AI sidebar to balance out colours and fix lightning in your images.

Tone enhance

Choose the Tone Enhance sidebar for higher dynamic range.

Color enhance

Adjust Color Enhance to increase saturation, and provide brightness adjustments

💡 Dependance - when to apply Light AI, Tone Enhance and Color Enhance

👉🏻 To prevent artificial lighting of your processed images after selecting Light AI, it is not recommended to apply Tone and Color enhancers
👉🏻 For Tone Enhance or Color Enhance, intensity levels of 15-30 will be good starting points, providing you hadn't previously applied Light AI. However, there's always a space to adjust their intensity further up to when the demanded level/satisfying outcome is achieved. Experiment!

Learn more: Tutorial #1: How to Upscale a Photo in One Minute with Let's Enhance

Up next: What are Presets and which one should I choose for my photos?

Updated on: 13/05/2025

Was this article helpful?

Thank you!