Inspiration
We are living in the changing world of COVID-19, and one of the most change that most of people face today is WFH. Working from home is a frustrated experience because home is not a suitable place for working, but relaxing and comforting, Changing workplace environment can lead to both physical health and mental health problem.
What it does
FastDoc is the fast and fun line chatbot that doing the daily check-up the employee. It collects the office syndrome data both physical and mental through text and video data, and analyst the employee health related to office syndrome. It help the company monitor and express their care to their employee by giving the right tools and supports.
How we built it
We build FastDoc interface using the line-sdk api and using the NLP of intent recognition to making a smart interaction with the user while collecting data with entity extraction. We build the computer vision to identify the neck pain of user through the deep learning of YOLO architecture and Headpose estimation to detect the degree of neck rotation, and use that to identify the symptom of neck pain.
Challenges we ran into
After we present the idea to the mentors, we gain the challenge of how can we show that this technology can potentially care for people who its monitor, we also finding the chatbot is not interactive enough and need to redesign and gamify so that it is appropriate for daily use.
Accomplishments that we're proud of
We successfully made the prototype that worked and create the great conversation with user, and we can create the model that correctly identify the rotation degree and the symptom of neck pain.
What we learned
We learn how to fit the right solution to the right problem using the technology that available and usable nowadays for the relevant problem that everyone face. The most value lesson learnt is how to build something with little amount of time but suitable and usable product.
What's next for FastDoc
FastDoc is planned to test the MVP with more user that gain a value feedback after that we will redesign and gamify our solution so that it is very fun and very easy to use. Later, we will go tackle the emotion part of office syndrome and going to be customizable for company by joining the Messaging api each company use
Built With
- computer-vision
- deep-learning
- dialogflow
- face-detection
- headpose-estimation
- line-sdk
- mood-detection
- natural-language-processing
- ngrok
- object-detection
- python
- yolo

Log in or sign up for Devpost to join the conversation.