Determining Scribble Intention in Hand-drawn Musical Notes

Sketch recognition is the area of artificial intelligence dealing with recognizing a user's intended meaning in hand-drawn sketches, as drawn with a pen on the screen of a tablet PC. Much work has been done to identify a variety of primitive shapes, such as lines, arcs, ellipses, and curves. Previous systems have also built scribble recognizers, but they have not examined the difficulty of identify user-intention of the hand-drawn scribble, as similar looking scribbles can be used in a myriad of contexts. Sheet music is one domain where a single scribble can be used to scribble out a note or fill it in to turn a half note into a quarter note. We are building a sketch recognition system to recognize simple sketches of hand-drawn sheet music that allows users to color in notes, creating half notes, as well as to scribble out notes, deleting them. We have performed a user study to identify features that help distinguish between these types of scribbling.

Poster Presentations

• Dahmen, K. and Hammond, T. Determining Scribble Intention in Hand-drawn Sheet Music. Grace Hopper Celebration of Women in Computing, Orlando, Florida. Oct. 17-20, 2007.
  
  
• Dahmen, K. and Hammond, T. Determining Scribble Intention in Hand-drawn Musical Notes. Industry Affiliates Program, Texas A&M University. March 7, 2007.