Neural Art: Introduction, Challenges, Implications

Neural art refers to visual art that is created at least partially by a neural network. While neural networks have gone in and out of favor among AI researchers over the last 80 years, neural art has recently surged in popularity due to advances that have enabled the easy generation of aesthetically pleasing and coherent visual outputs. We are at the precipice of paradigmatic change, both within the art community and in any domain that visual media touches. As we hurtle toward a strange new future for art and creativity, it is important for us to collectively shape it with a consciousness of the values being embedded in it. However, there is a lack of discourse in this area. This thesis is an attempt to introduce the concepts behind neural art and the concerns surrounding it, as a jumping-off point for increased understanding, discussion, and collaboration. Writing for a non-technical reader, I start with a primer on neural art. Some techniques are further developed in five personal works in the next chapter. This is followed by findings from a 43 participant survey regarding societal concerns about neural art, and a discussion of pertinent latest challenges. Based on the research, ten implications are discussed. The most updated version of this manuscript can be retrieved at https://github.com/oliviaseow/pixels2picasso.

Name thesis:degree_name

Master

Department dc:contributor.department

Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2022

Author dc:creator

• Seow, Olivia

Advisor dc:contributor.advisor

• Mueller, Stefanie

Statement dc:rights

• In Copyright - Educational Use Permitted
• Copyright retained by author(s)

Licence dc:rights.uri

• https://rightsstatements.org/page/InC-EDU/1.0/

Handle dc:identifier.uri

https://hdl.handle.net/1721.1/145103

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/145103