FolkFinder

Inspiration

We were primarily motivated by how old people may often get lost. In one local story, an old man passed away unnoticed and his body was only found days after his family notified the authorities. This brought up anxiety amongst some of the locals who believe in an urban and developed city like ours, we could do better for these individuals.

Based on this, we were motivated to develop a solution to ensure that

What it does

The system detects when a person enters a location and keeps track of how many people remain at the location, if an individual has remained in there for a long time, the system will sound an alarm and drop a message to the management of the place

How we built it

For our model, we utilised the capabilities of an existing object detection library to suit to our context. We edited a few lines to capture when the target has enter/ is leaving the scene and then adapted it with a raspberry pi. The raspberry pi informs the relevant user whether the target is within sight through the use of a telegram bot, and also LED and buzzer.

Challenges we ran into

We had initially planned to use an arduino instead of a raspberry pi, but due to the wifi shield not working. Too much time was spent on debugging the shield

Also, we were rather new to the ML/AI scene. Not many of us knew of ML/AI prior to this, so we just wanted to give it a shot and understand some of it through this hackathon. It was a challenge to find the right repos and code to suit the context, as well as adapting it.

One other thing was how to coordinate our efforts. For that, we tried github, but we also did not know how to use it thoroughly, so we experimented some methods in the process of doing the project.

Accomplishments that we're proud of

• It works (although incomplete)
• We did this while learning new stuff concurrently

What we learned

• How to use Raspberry Pi GPIO
• How to program a telegram bot
• How to implement a Deep Neural Network (DNN) based object detector and tracker
• Better understanding of GitHub and version control
• OpenCV stuff

What's next for FolkFinder

This prototype only showcases what would happen if a person was detected within the frame, but more can be done to show whether the person is indeed the intended target/if the person is accurately leaving/entering the room or is experiencing difficulties/ just interacting as per normal. More can bedone to address these problems before the inspiration can be realised.

Built With

• c++
• computer-vision
• deepsort
• iot
• python
• pytorch
• raspberry-pi
• telegram
• yolov5