Sign in to follow this  
  • entries
  • comments
  • views

Computational Aesthetics, part 2

Sign in to follow this  


Today was another full day of sessions. I'm providing my understanding of some of the sessions below; there may some simplifications or misunderstandings.

David Mould described a method of placing stipples to maintain the visibility of the edge between colors in "Stipple Placement using Graph Search". He used a pathfinding algorithm with a cost function that took into account contrast as well as value. His algorithm was localized in that each stipple was placed relative to the last stipple, so that the image was built up from an arbitrary starting point. The results did not exhibit the regular dot patterns found in other stippling/half-toning methods.

David Sprague talked about personalized customization in "Individual realities: Customizing Aesthetics in Shared Immersive Virtual Environments" (Sprague and Amy Gooch). He suggested presenting special representations of virtual spaces to each collaborator depending on their needs. The idea I took away from this was the possibility of using a fragment shader pass over a scene to remove a color and test the accessibility of a game for color blind people. This could be added as an additional command-line option for PAGE pretty easily.

Mateu Sbert presented "An Information-Theoretic Ambient Occlusion" (Francisco Gonzalez et al). He suggested selecting the best n light sources from a regular set of potential sources surrounding an object. If I have this right, the best source is selected based on the number of pixels rendered over the total that would have been rendered without occlusion for all polygons. I suppose ARB_occlusion_query would be useful here. Then its a simple matter of rendering with all selected light sources, adding them up and normalizing the result. The more light sources the better the result. He also did some colorization by assigning a color to each light source.

Tomorrow is the final day, ending at around 3:30 or so. I'm looking forward to getting back and perhaps trying some of these ideas in my own work.
Sign in to follow this  


Recommended Comments

There are no comments to display.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now