← All guides

Who we are and what we do

CraterView.ai is a Canadian startup building image enhancement and restoration models.

We want to make the state of the art in image enhancement available to everyone. In practice that means three things:

1. State of the art. We follow the field closely and do our own research and development to bring the best of it into our models.
2. Easy access. We run the models, so a browser or a phone is all you need.
3. Community. Our models are free for casual use, on our community queue.

How it works

You send a picture, choose what you want done, and get the improved image back. You can use our web app, our developer API or let your AI assistant handle it. Our iOS app is coming soon.

Main models:

• Enhance and upscale: enlarge an image and recover detail.
• Denoise: take the grain out of photographs shot in low light or at a high ISO.
• Restore: repair scratched, faded or damaged prints and old photographs.

Our NSFW content classifier and our real vs. generated classifier are available via the web app and API and we’ll aim to keep them free. From time to time we’ll also release fun enhancement models from our lab, for headshots, image filters and more — try the starship captain look in Headshot if you haven’t yet :-). Our main focus, though, is image and video enhancement and restoration.

Where it helps

Outside of the lab, we use the models while on hikes and other backcountry adventures to bring details closer, see further and capture sharper images. Have a look at the gallery (once you’ve signed in) for how others are using our model, and at our guides for interesting use cases, tips and tutorials.

Why now

It is a great time for vision technologies. In just a few years, image models have gone from smoothing pixels to understanding what a picture shows. That understanding lets us enhance and restore images in a way that is more faithful to the original scene.

The catch is access. The best of this work lives in research papers, needs large graphics cards to run, and takes real expertise to set up. Most people who could use it never get to.

Where to start

• Fix a photo: open the web app and upload one.
• Build with it: your first API call takes three requests, in curl, Python or TypeScript.
• Work through an assistant: connect one to CraterView, or install the skills.

More guides

Basics API Photography Bit depth, colour profiles and file formats Bits per channel, transfer curves, colour profiles and phone HDR, what each format holds, when not to use auto, and what to keep for the web, print, an archive and science. Basics Credits and subscriptions explained How the free community queue works, and what credits and subscriptions add when you need more than it gives.