Hey

Inspiration

Our inspiration came from a mix of previous experience with computer vision and recognizing how it can enhance the shopping experience and services. Understanding that sometimes customers may find it difficult to locate an employee, we decided to let customers notify employees by simply raising their hand.

What it does

When a customer raises their hand, employees will be notified of the customer's location and be able to arrive at their location to provide assistance.

How we built it

The cameras recognize customers raising their hands by machine learning via Tensorflow. A signal is then sent from each camera to a server, indicating whether there are people requesting help in an area. States of each area are then broadcasted to each employee to notify them of their requests.

Challenges we ran into

Understanding Socket.io for the first time. Setting up JavaScript and Python Clients with a JavaScript server.

Accomplishments that we're proud of

For some of us, developing a project using tools we haven't used before such as Socket.io and Javascript. How to set up a Server back-end. For most of us, the first Hackathon we competed in and first productive all-nighter.

What we learned

We learned how to collaborate effectively. We are used to coding solo, so it's not common for us to work with other programmers on a project. With the use of GitHub, we can keep track of each other's progress and improve our effectiveness in coding.

What's next for Hey

The current prototype is still quite primitive. While it can be implemented with some success, there is still a lot of improvements to be made. One such example is to check the availability of an employee and deciding whether or not to send them a notification. Also, we can partition aisles by recognizing shelves and reducing the cameras needed, hence reducing operational costs. Multi-person tracking is also a goal for the betterment of this project.

Built With

• javascript
• opencv
• python
• react
• socket.io
• tensorflow