Written on September 16, 2008
Golan Levin’s paper titled Computer Vision for Artists and Designers provides a nice introduction to computer vision using historical context, introducing computer vision in art and also providing a semi-technical perspective on what computer vision is and the techniques it employs. For those of us new to the field, it provides a nice foundation and even goes as far as providing a starting point (including code) for our own explorations.
This piece connects directly with the theme of this course and provides food for thought in the melding of art and computer vision. Computer vision as an input shifts time, objectivity, abstraction and also provides automation. With this as a tool artists now have a new level upon which to build from. While the content can vary, computer vision does provide a new way to gather and process visuals and make the viewer an active participant in the experience. Given this type of interaction the viewer is not only a subject IN the piece (as in a portrait, which captures an instant of time) but may also influence it on a continual basis. Working with computer vision as an artist, a different type of abstraction must be employed and one would hope that new creative avenues are uncovered. Similarly, working with computer vision as a developer WITH artists one would hope that a shift in perspective would also help innovation, after all software development and art gain a great deal from the practice of abstraction.
In addition to OpenCV, there are also various projects working on the field of Computer Vision. One such project is Torch3Vision, found here. Although the field of computer vision has been around since roughly 1966, it is constantly and rapidly evolving. As the tools and quality of computer vision continues to improve it will be interesting to see how artists use computer vision and possibly drive it in new directions.