-
-
Face detection
Inspiration
In the book 1984, Big brother is always watching, privacy is no more, we wanted to create something similar to big brother. We have a scoring system, where people have a score, this score is based on their behaviour, points are decreased or increased depending on the behaviour, this was taken from a Black Mirror episode.
What it does
A camera detects peoples faces, it then watches their actions, if the action is bad, their Big Sister score decreases, and a voice tells them to stop, if the action is good, their Big Sister score increases, the a voice tells them to keep up the good work. There is a ranking system, there are thresholds which people can reach, that will allow them to use public transportation and more basic necessities.
How we built it
The face detection was built using open CV. Gemini is used to detect user behaviour. Neuphonic was used for text to speech, using the output of Gemini to communicate with the user of the program.
Challenges we ran into
Connecting gemini api to open CV was difficult, the LLM can't return structured output + context to interact with the face detection system in one query.
Accomplishments that we're proud of
Even though we were a team of 2, we did everything that we wanted to do.
What we learned
We learned the power of teamwork, and effective collaboration. We also learned how to prompt gemini api to give a structured result
What's next for Big Sister is Watching
We hope big sister ceases to exist, we made this demo to show how bad the future can be if privacy is no longer.
Log in or sign up for Devpost to join the conversation.