Articles on: Getting Started

Which operations to choose?

Choose the right algorithm best suited for your use case



Enhance photos taken by phone, digital camera, small compressed & pixelated, for printing, photos of people, architecture, cartoons and more!

Let's Enhance algorithms work with almost all types of photos: from user generated content taken with a mobile phone, to professional content captured with digital cameras; low resolution, compressed & pixelated images are no longer a problem thanks to our advanced algorithms designed to detect and eliminate noise and artifacts, in just one click. Enhance your digital art (illustrations, anime, cartoons) by simply applying pre-defined upscale types. With our Web Platform, images with captions get more detailed information, and it's up to you if you want to edit maps, logos, posters, etc. Need to remove background? Let us do that for you too! 🤗

To do all this, we've developed different OPERATIONS where you can adjust the upscale type, output size or enhance tone and color according to your current needs, and PRESETS packed with predefined solutions that improve your images for printing with one click.

Our platform applies a noise reduction system whenever you upload images with the .jpg or .jpeg extension, as our algorithms automatically identify them.




Here's a quick guide of what choices you have when it comes to enhancing your photos!

OPERATIONS







Within this tab, you will find:

Upscale type


Smart Enhance

Smart Enhance specializes in low-quality images (compressed, blurred, pixelated, etc.).

Works on low-quality photos of nature, architecture, products, real estate, and food, enhancing clarity and detail
Increases resolution by 1x, 2x, 4x, 8x and 16x factors

However, this neural network cannot recognize images with bodies. It is better to use the 'Old Photo' or 'Old Photo plus Enhance Face Details' upscaling types for this particular use case.


Digital Art

For all types of digital art, including illustrations, cartoons, and anime
Allows you to increase your image resolution by 2x, 4x, 8x and up to 16x


Image with Text

Designed to enhance images with all types of text, such as labels on product packaging, posters and maps.
Provides smooth results
Increases image resolution by 2x, 4x, 8x and 16x factors


Old Photo (Legacy Photo upscale type)

For early digital or noisy photos
Suitable for low-quality images of all types (people, nature, architecture)
Gently smoothes noise with minimal risk of artifacts
Upscales up to 2x 4x 8x 16x factors
Allows manual resizing (preserving the aspect ratio of the image)

Using the Old Photo algorithm is not recommended for good, high quality images because the upscaler type 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.


Old Photo + Enhance face details 

Together, these two algorithms do an excellent job of enhancing photos with faces, removing artifacts, redrawing missing elements, and adjusting the sharpness of each line to produce sharp and clear images.

'Enhance Face Details' is a separate AI algorithm trained on millions of face datasets, so this neural network consistently detects missing elements and determines what needs to be adjusted.
Upscales up to 2x 4x 8x 16x


Magic (beta) 

Next-generation, most powerful AI upscaler for AI art, food and landscape photos, and images where fine detail isn't critical
Works with and without prompts
Allows you to set the creativity level (the higher it is, the more ingerention in your original image, we suggest between 10-40%)
Increases image resolution by 1x, 2x, 4x, 8x and 16x factors
Allows manual resizing (preserving the aspect ratio of the image)

Advanced options



Remove background

Removes the background from your images. Remember to set the output file format to PNG before processing. Here's how.

Size

Allows to set your custom values for width & height and resolution

Light AI

Neural network that enhances the light (exposure), colors (saturation), depth (contrast), and detail

Tone enhance

Apply for higher dynamic range

Color enhance

Increase saturation and provide brightness adjustments



💡 Read more which algorithm to apply in your case: What operations work best for different types of photos?


PRESETS



Printing


It’s a combination of the ‘Photo’ algorithm and a variety of output formats for printing purposes (such as posters, photos, and international papers); however, it’s crucial to keep in mind, that it does not work well with blurry images.









Up next: What type of images I can use with Let’s Enhance?

Updated on: 17/07/2024

Was this article helpful?

Share your feedback

Cancel

Thank you!