Pixelbot 3000: Turn AI ideas into Lego puzzle masterpieces

An enthusiastic YouTube user has designed and built a Lego printer that automatically assembles complex Lego puzzles, like Lego art sets featuring da Vinci’s Mona Lisa or Hokusai’s The Great Wave. This machine is inspired by another Lego printer called Bricasso , but it uses artificial intelligence (AI) to simplify the process.

Although Jason Allemann’s Bricasso was technically impressive when it launched eight years ago, it required a complicated process: users had to create their own mosaic design, print it on paper, and then have it scanned by a scanner. image. YouTube channel Creative Mindstorms used some custom code and AI so users just need to enter a simple request.

With Pixelbot 3000, users simply enter the artwork they want the printer to create. This command is sent to OpenAI’s DALL-E 3, with a request to create a simple animated image, sized 1024 x 1024 pixels.

This printer can only fit much smaller mosaics, consisting of just a 32 x 32 grid of Lego bricks, but instead of shrinking the image created by the DALL-E 3, the Pixelbot 3000’s code will split the AI-generated image into a 32 x 32 grid and sample the color of the center pixel in each square. This helps create high-contrast miniature images, resulting in a better-looking mosaic.

Another limitation of using Lego as an art medium is that the plastic pieces only come in about 70 different colors, and the Pixelbot 3000 only uses 15 of them. The AI-generated thumbnail image is subjected to a final processing step to find the best color match for each colored pixel, corresponding to the 1 x 1 Lego bricks used to assemble the final puzzle. .

Designing, building and programming the Pixelbot 3000 seems to be as labor-intensive as assembling a Lego puzzle, which can have more than 11,000 pieces. If you really want to reduce stress, just spend 15 minutes watching the Pixelbot 3000 in action, witnessing the process of turning ideas into reality.