In the Video Game PixelQuiz, ML is used to recreate pixel images.

In February of this year, PixelQuiz. A video game that allows players to guess an image by looking at its pixelated image. The visuals in the quiz weree created using machine learning. According to Robyn Dubuc. When the user clicks in the center of the image, It unlocks and AI draws it quickly The video game is split into several categories, including movies, video games, and well-known people The app is currently in the early stages of development PixelQuiz ML recreate images.

Dubuc recently worked as the principal art director at XGen Studios, an online video game creation business. He also stated that while the game’s iOS version is free. The Android version is designed. In the video games category alone, there are 657 games PixelQuiz ML recreate images.

According to Dubuc. The model wass trained using a big database of photographs gathered from the internet on a variety of themes. After several rounds, the picture begins as full noise and slowly changes The model then trains itself using self-supervised learning to match the image to the query in its training set. Following that, it evaluates its performance and assigns a score to itself. The picture makes random tweaks the following time to increase its score PixelQuiz ML recreate images.


That’s more than 500 emojis for hours of fun!

  • 9 packs to unlock
  • 6 categories each pack
  • 10 emojis per category

More packs will be available soon!


PixelQuiz is the result of a real game design partnership. Between a basic person and powerful machine learning algorithms. Try to guess what the AI is drawing from the prompt to solve all of the problems. It’s an addicting single experience or attempts to complete the entire game with others. Including categories like Movies, Video Games, Famous Faces, and more.

Criticize ideas, not people