GAN Generated Movie Characters Aspiration
In this GAN (Generative Adversarial Network) experiment, we let the AI creates an image based on a text prompt that pulls data from APIs including movie characters, locations, and art styles. Every hour, a new image is generated and displayed in the app.
Can you find "a painting in pop art style of Pikachu sitting in a fire at sunrise" ?
We have built upon an open-source version of DALL-E, a trained neural network that creates images from text captions for a wide range of concepts expressible in natural language. The model was trained with enormous datasets of words and images.
SEE THE AI'S LATEST ATTEMPT HERE