Observations

One of the most interesting observation we made is that the newest LEAP SDK version provided little to no improvement over the previous version. While they claimed improvements in their tracking algorithms, we saw equivalent results in our own usability studies.

Most of the project was very straightforward to implement, but took a long time to tune. The bounds on each gesture implemented took extensive trial and error to get right and, even then, one of our gestures still performed terribly in usability studies.

Future Work

The most important next step is to expand compatibility to more window managers. As far as code is concerned, the calls are easily swappable and if the targeted window manager supports the necessary calls, support should be trivial.

It would also be great to provide a user interface for the configuration. XLeapr is meant to target average users. While we felt a UI was out of scope for the project, a configuration file does not meet this goal. We envision a UI which allows the user to easily change gesture bindings, as well as download new gestures from a sort of developer marketplace.