Related Work

Table of Contents generated with DocToc

• Summary
• [MAGIC](#magichttpdanielashbrookcommagic-a-motion-gesture-design-tool)
• [Quill](#quillhttpciteseerxistpsueduviewdocdownloaddoi1011242009&reprep1&typepdf)
• [Gestalt](#gestalthttpresearchmicrosoftcomappspubsdefaultaspxid141330)

Summary

Year	Paper	Sensor	ML Algorithm
2001	Quill	pen stroke	Specialized
2010	MAGIC	accelerometer	DTW

MAGIC

Quill

quill supports gesture creation by example, and offers automated advice on improving the gestures.

Lesson Learned:

• some users did not realize that two similar gesture classes might conflict with each other
• users didn’t understand how the gesture recognizer worked
• and finding and fixing problems with recognition was very difficult

To help alleviate these issues:

• quill includes feedback on the quality of gestures and used language to communicate issues

However, "many participants did not understand the suggestions," going on to posit that perhaps they "can be made more accessible. . . by using more diagrams."

Gestalt

Kayur Dushyant Patel's thesis has more related interesting information.