omniVision

See our slides!!

https://docs.google.com/presentation/d/1YA8PM-_N5l7qRW6BOi42NMncI9d9-h_7uI2vCf8yuLQ/edit?usp=sharing

Inspiration

Home security is a very important aspect of our life. When we are away for a long time (say a family trip during summer vacation), we can't always keep an eye on things at home, so this smart application promises to do the heavy lifting for us.

What it does

Project omniVision not only provide surveillance camera coverage over our house, but also is capable of sending us real time notification of motion anomalies within our home with the frame and the intruder’s face captured.

How we built it

The omniVision project consists of three major parts:

• A real time camera feed analyzing script implemented with machine learning and computer vision algorithms
• Dropbox and Firebase as a quick means of storing captured frames and send notification to the application.
• A front end web app that can be opened on many portable platforms and interfaces.

Challenges we ran into

• Constructing the machine learning and computer vision algorithm and calibrating it for our purpose.
• Binding the three parts together was a big problem.

Accomplishments that we're proud of

• Binded everthing together with the api provided by Dropbox.
• Working machine-learning and computer vision algorithm!!

What's next for omniVision

• Communication between multiple cameras

Built With

• computer-vision
• dropbox
• firebase
• javascript
• machine-learning
• opencv
• python
• raspberry-pi
• rest-api