Project

Paper Dreams

Paper Dreams explores the various modes of creativity that can be enabled by artificial intelligence.

A fair amount of current research has been focused on using machine intelligence to learn “creativity,” from transfer learning of artistic styles to translation of languages. However, the application of current machine learning algorithms and multi-modal inputs to the augmentation of existing human creativity is a relatively unexplored area. By focusing on the dynamics of this human-machine interaction and working with the representations inside machine learning models, we can give people new tools for reasoning.

Our system at its current form provides a user with a canvas that it is more like a mirror, where we draw strokes and re-evaluate what our drawing looks like at each step in order to continue. We do this by building a neural network which takes a small number of input variables, called latent variables, and produces the entire sketch as output.

The machine can use analog modes of thinking to recognize what we are drawing and help us move forward. At the moment, Paper Dreams augments the drawing experience in thre… View full description

Paper Dreams explores the various modes of creativity that can be enabled by artificial intelligence.

A fair amount of current research has been focused on using machine intelligence to learn “creativity,” from transfer learning of artistic styles to translation of languages. However, the application of current machine learning algorithms and multi-modal inputs to the augmentation of existing human creativity is a relatively unexplored area. By focusing on the dynamics of this human-machine interaction and working with the representations inside machine learning models, we can give people new tools for reasoning.

Our system at its current form provides a user with a canvas that it is more like a mirror, where we draw strokes and re-evaluate what our drawing looks like at each step in order to continue. We do this by building a neural network which takes a small number of input variables, called latent variables, and produces the entire sketch as output.

The machine can use analog modes of thinking to recognize what we are drawing and help us move forward. At the moment, Paper Dreams augments the drawing experience in three ways: by adding textures/colors, suggesting other elements/drawings for the scene, and introducing serendipity. To adjust the level of serendipity, the user has control of a dial that determines how "predictable" vs "unpredictable/nutty" you want these machine additions to be.

Some of the driving questions for this project are:

  • To what extent are these new tools enabling creativity?
  • Can they be used to generate ideas which are truly surprising and new, or are the ideas clichés, based on trivial recombinations of existing ideas?
  • Can such systems be used to develop fundamental new interface primitives?
  • How will those new primitives change and expand the way humans think? 

Paper Dreams Team:  

Guillermo Bernal, Haripriya Mehta, Lily Zhou