Elvira Smit

From sketch to (weird) photo. How?

Sunday 19 April 2020, 15:51 by | 4,744 views | 0 comments

This article was translated from Dutch with the help of AI. Read the original.

Looking at Bob Ross, I think, 'I can do that too.' But the result is usually a pitiful little work that looks like nothing. That no longer matters, because with a bit of endless sketching I can now conjure up a masterpiece in no time.

A computer hardware producer, NVIDIA, has developed something using a self-learning algorithm. Draw a few lines and the algorithm turns them into a photo. A green area becomes grass, a few brown lines become a little house.

You feel just like an artist. The name of this programme also winks at that, with the name GauGAN. Although the painter Gauguin had to put considerably more effort into a little landscape.

From sketch to photo


The algorithm updates the photo every time you add something new to the image. If you sketch a little lake, or a simple blue area, and later add some trees along the bank, the reflection in the water changes too.

In short, you can make a sketch at the level of a toddler, a house, a tree and an animal, after which the algorithm fills everything in and turns it into a realistic photo. The shadows, lighting and season are all correct. A flowering tree in a snowy landscape does not occur.

Of course, I immediately started colouring myself. I had a good half hour of fun with it. It takes a little figuring out how exactly the algorithm 'thinks', but you discover that as you play. And if you don't like it, you can start again endlessly.

Rocks in the sea look like this:

Another sketch


These are not combined photos


My result may not blow you away, but the results are getting better all the time because there is a self-learning algorithm behind it. The same company previously made a product that now produces lifelike portraits all by itself.

That also started with many flaws in the results. But just a few years later, the photos are eerily real. See for yourself.

A while ago now, in 2015, I wrote about an algorithm that recognises beautiful photos. That naturally raises a lot of questions, because who decides what is beautiful?

GauGAN bypasses that question: it fills in a sketch based on millions of photos it has seen itself. So actually just like the painter Gaugain, who created a unique painting based on all the images he had ever seen.

Sketching yourself?


If you also fancy doing some sketching, you can do that here. But you can also, even more lazily, watch the video below and see how others cobble all sorts of things together.

Share: WhatsApp Facebook
Elvira Smit

About the author

Ik heb een eigen fotografieblog, Kronkeling. Daar schrijf ik om de week twee artikelen. Over hoe je je blik kunt scherpen, je creativiteit kunt aanwakkeren, en over fotografen die mij inspireren. Geen technische zaken, wél over hoe je leert kijken. En dat blijkt nog een hele klus. Een mooi startpunt zijn de Kronkeltips, die vind je hier.

Show every article in Oddities

0 comments

    There are no comments on this article yet.
    Be the first to share what you think!

Have your say

Please note: Articles older than a week can only be commented on by signed-in visitors.
Would you still like to comment? sign in or create a free account.

Become a better photographer every week

A practical photography tip in your inbox every week.
38,982 photography enthusiasts already receive the tips! More about the weekly mail. Or keep up to date via Facebook. Sign-up secured with Google reCaptcha.

Elja Trum

Elja Trum

Photofacts; everything to do with photography!

Would you like to take better photos and keep up with what is happening in photography? Photofacts publishes articles that teach you something about photography — from a product announcement to a practical shooting tip or a look at a website. Photofacts covers photography every day and is a hobby project of Elja Trum's that got out of hand. The articles are written by a team of volunteer authors. If you would enjoy writing one or more articles for Photofacts, do get in touch.More about Photofacts

Photofacts is made possible in part by


We use cookies to make Photofacts work properly. Read our cookie policy. got it