Projects‎ > ‎

Opportunities for Students

This page lists some of the projects for which I am currently looking for students to work with me on.  All of these projects involve the LogiSketch system my students and I developed at HMC, so please read a little about the LogiSketch system first.   I don't have much funding (yet) but these would make nice Master's projects or theses for students who are looking for those.  Please contact me if you are interested.  I'm happy to talk to you more. 

Also, if you are interested in tutoring for me, please see this page.

User study to investigate how to influence drawing style

I am looking for one or two students to continue a study that we started at HMC a year or so ago.  The idea is to investigate recognition feedback mechanisms that can (potentially) influence users to change their drawing styles so that the system recognizes their drawings better.  The work will involve working with the LogiSketch system we developed at HMC and modifying and re-running a user study that was designed and run by students at HMC.  This is a perfect opportunity for students interested in user interface design, particularly intelligent or recognition-based user interfaces.  It's no problem if you don't have a Tablet PC.  I have several available for you to work with.  Some experience with development on Windows and some familiarity with user studies is a plus, but not necessarily required. 

For more information about LogiSketch, please see my sketch-based interfaces project overview page

LogiSketch development and deployment

At HMC my students and I developed a sketch-based interface for drawing and simulating digital logic circuits.  More information about this project can be found on the sketch-based interfaces project overview page.  I am looking for someone to continue development of this tool and to deploy it with UCSD students, either formally or informally.  This work will involve porting the code repository onto a publicly accessible site from its current location at HMC, and testing and updating the system to be more reliable.  It will involve work developing and improving both the existing recognition algorithms, but mostly the user interface.  The system is implemented in C# on a Windows Tablet PC.  Some experience with large-scale Windows-based development is a plus, but it's OK if you don't have specific experience developing on a Tablet PC or with C#.  I have Tablet PCs you can use.  

Sketch Recognition Algorithms

One of the most challenging aspects of developing a sketch-recognition interface is to develop recognition algorithms that are robust enough that users will accept them.  Although our LogiSketch tool often performs quite accurately for some users, it is not yet robust enough to be practical.  I am looking for someone who is interested in designing and implementing algorithms to improve sketch recognition in the digital logic domain as well as in other domains.  I have plenty of data for you to work with.  Your goal is to research, design, implement and test algorithms that will improve recognition.  This is primarily a pattern recognition/machine learning project.  Some familiarity with basic machine learning and/or pattern recognition is required.  I have Tablet PCs you can work with, though working on a Tablet PC is not required for this project.