Make-A-Scene is an artificial intelligence (AI) research idea that Meta is showing off. It will let people bring their ideas to life.
People can make photos with Make-A-Scene by following text prompts and drawing freely. Before, AI systems that made images usually took text descriptions as input, but the results were hard to predict. For example, the text “a painting of a zebra riding a bike” may not be exactly what you had in mind. The bike may be facing the wrong way, or the zebra may be too big or too small.
This is no longer true with Make-A-Scene. It shows how people can use words and simple drawings together to get their ideas across more clearly by using a variety of tools.