Visionle

Visionle is a visual word-guessing game where players must assign labels to images drawn at random from the ImageNet-Sketch dataset. Combining image recognition with word-guessing mechanics, the game offers players a fun and challenging experience.

How to play

The primary objective is to accurately predict the label or phrase that best describes a given image from the ImageNet-Sketch collection. Players have up to six tries to determine the correct label for the image. In each trial, you had to guess which label went with the picture that was shown. In this case, they can use any letter combination that is less or equal in length to the real challenge word. A more laid-back "Looney Mode" is also available, in which players can make a wide range of guesses. After entering a guess, the game will provide feedback based on how well you predicted. Cells containing the letters of the word change color to reflect the following:

The participant keeps making educated guesses based on visual cues from the image and using the feedback they receive to learn the proper label. Winners are those who correctly estimate the label within the allotted six tries.

Try more Wordle games at the latest Wordleverse collection!