qcards.txt

qCards is a low cost audience polling system using computer vision. It uses a camera to identify patterns printed on A4 size paper and log the responses. This was my internship project at <b>Microsoft Research Lab, Bangalore, India</b>.</br>Electronic clicker devices are widely used for attendance, polling and quizzes. But the problem with these clickers is there cost. Each clicker costs around $30 and then there is cost associated to the response logging system that summarises responses from all the devices.</br>The huge cost involved, restricts the use of such system only to organisations who are financially well-off. Thinking in the context of a developing country like India, such sytems are not affordable for a mid or low income level school classroom. qCards addresses this problem and provides and alternative.</br>The system consists of a camera and several A4 size papers with patterns printed on them(the cards). Each member in the audience is given a card corresponding to his/her id. Each of the 4 possible orientations(obtained by rotating by 90 degrees) of the card codes for A, B, C or D. The participants lift their cards with the pattern facing towards the camera in front. The patterns are then captured and processed.</br>I developed a complete Android application that uses OpenCV4Android for image processing tasks. The app has several features which include attendance and quiz conduction, roster management and quick poll.</br>You can find more about this project <a href="http://research.microsoft.com/en-us/projects/qcards/">here.</a></br>I'll upload a demo video soon!