Comment Location:
http://martysimpossibletorememberurl.blogspot.com/2010/09/reading-2-rubine.html#comments
This is the paper that virtually started sketch recognition. The author bundled up his research into a sketch recognition program called GRANDMA. After preprocessing, the gesture is used to derive 13 features. The author mentioned the feature set could not distinguish between all types of gestures and that the feature set should be expanded in future implementations. The technique employs machine learning and a linear classifier.
The author allowed for a classification to be unknown. The thresholding measure utilized would sometimes reject known gestures; obviously, future improvements are needed in that area (it's probably already been done).
Discussion:
Since there are many more sophisticated sketch recognition programs in existence today, I did not consider the GRANDMA program to be of great significance outside reference material for building one's own sketch recognition program. The preprocessing seemed rather arbitrary (only consider every 3rd pixel or so); I'm sure there are more effective preprocessing measures in existence today. Concerning the unknown option for classification, I'm certain the author did that to allow for future improvements.
I also thought it was interesting that he skipped to at least every 3rd pixel. But what where the standard monitor resolutions back then? Maybe that had something to do with his choice.
ReplyDeleteI think one has to consider that this work was done almost 20 years ago and is probably the foundation for a lot of the sketch recognition programs in existence today. Even though I'm sure that GRANDMA has since been improved upon, it was still successful in allowing users to create gesture recognizers which is significant.
ReplyDeleteI think IBMs XGA supported 1024 x 768 in 1990.
Dean Rubine here. It's been a while, but I don't recall skipping pixels except for the features that determine the initial angle of the stroke. When people would make strokes with the mouse it would sometimes take them a bit before they got going in a clear direction, so just using the first couple of points often gave a bad answer. My computer was pretty slow (a Microvax II, probably vintage 1984) so the mouse points didn't come that frequently. If I was doing it today, I'd probably wait a certain length of time before determining the initial direction of the stroke.
ReplyDelete