“Quick, Draw!” is an online game developed by Google that uses artificial intelligence (AI) to recognize and classify drawings made by players. The game works by presenting players with a prompt, such as “Draw a washing machine”, and then giving them a few seconds to draw a representation of the prompt using the mouse or touchscreen.
As the player draws, the AI system uses machine learning algorithms to analyze the lines and shapes being drawn in real-time and classify the drawing as a specific object or concept. Once the player is finished drawing, the AI system displays the top three guesses for what the drawing represents, along with the percentage confidence it has in each guess.
“Quick, Draw!” was developed to demonstrate the capabilities of machine learning, neural networks and NLP and to help people understand how these technologies work. The game has been played millions of times by people worldwide and has helped to popularize the concept of machine learning and AI.
AI technologies used in this experiment are as follows:
Machine learning algorithms are a type of AI that allows computer systems to learn and improve their performance over time without being explicitly programmed. In the case of “Quick, Draw!,” machine learning algorithms could have been used to analyze the drawings made by players and classify them as specific objects or concepts.
Neural networks are a type of machine learning algorithm that is inspired by the way the human brain works. They comprise interconnected “neurons” that can process and analyze data and are often used for image and speech recognition tasks. It is possible that neural networks were used in “Quick, Draw!” to help the AI system more accurately recognize and classify drawings.
Other AI technologies that might be used in “Quick, Draw!” include natural language processing (NLP) for interpreting and understanding the prompts given to players and computer vision techniques for analyzing and interpreting the drawings.
Explore it here: Quick, Draw!