phantom

Inspiration

A friend of one of our group member's had a broken foot while having to give a presentation in university. Since they did not have an clicker to go through the Powerpoint slides, they were effectively confined to the area next to their computer when they presented. We quickly realized that there are many applications that a Powerpoint aid can help.

Unless the presenter has a remote, they are confined to the area around their computer when they want to change their slide. We aimed to create a Powerpoint aid that lets presenters feel comfortable moving around so that they can engage their audience better.

What it does

The user can input hand gestures into a camera to move their slideshow forward, backward, or escape from full-screen. They are also prompted to input a list of keywords that, when the user says them during the Powerpoint, will move the slideshow forward.

How we built it

The gesture recognition system is built using Python's OpenCL API, and the audio recognition system is built using PyAudio.

Challenges we ran into

Combining both the audio and gesture recognition system; when we built them, we built them in separate groups, and had to find a way to integrate them together.

Accomplishments that we're proud of

Using a technology (Python) that we are unfamiliar with, and it's frameworks, proved to be a challenge. Thus, being able to successfully run all of the components is an accomplishment we are proud of.

What we learned

We learned how important it is to look at your program development holistically; there are many gears in the machine, and understanding how it all fits together can ease growing pains as the program becomes larger and multiple components need to be combined.

What's next for Phantom

The primary limitation for Phantom right now is the hardware: since the webcam and the microphone are not professional-grade, they do not work as effectively as they might with better equipment. In addition, the gesture recognition system can be improved so that it recognizes when an object other than a hand is in the camera's view and to not base any gestures off of that.

Built With

• python