Inspiration
Our inspiration stems from the "Who are you?" (problem statement 1) question of Children's health. As patients, who would not want to know who all visited them and who all they missed seeing when they were recovering? Not only the cared ones, but also the empathetic nurses, the children bond with them pretty effortlessly. The idea we are providing is to make the lives of children easier in this social manner.
What it does
We came up with a facial recognition system, which recognizes the staff members of that hospital and automates the logging of entries of each visitor of that patient. These entries are marked with a friendly user experience using React, where a patient can drag, click and see each visitor and something quirky about them; just like the characters from any video game. If a staff member is unrecognized, or a distant family member visits, our recognition system asks them to identify themselves and sends an automated text to the patient's parents. This way, the visitors' convenience, and parents' apprehension are taken care of.
How we built it
We divided the whole project into 5 segments. Shreyans & Kruti - They gathered the required libraries and datasets (photos) for facial recognition, and enabled the very foundation of this project. We are using the face-recognition library which finds similarities in faces based on certain hyperparameters. We have tuned them for adequate accuracy but for minimum false positives. Rasesh - He made sure that the user interface is interactive and enough engaging for any child. He used bootstrap framework and vanilla Javascript for aesthetics and essential logic. Karan - He owned the middle ground - web development server. While handling the crucial transitions from the backend to the front end, he made sure that the recognition is conveyed fast, even when this solution is scaled, and that the latency is kept in check. Jayashree - She upgraded the project from a web application to a mobile app. With increased mobility, we are planning to add features like "order food" in the mobile app.
Challenges we ran into
Bringing every segment into sync was the biggest challenge. Even if any recognition is done, if that is not conveyed correctly and efficiently to the front end, it serves no purpose. Making sure that nothing breaks in this transition was fun to overcome.
Accomplishments that we're proud of
We learned a week full of technical learning in just 20 hours. So much current-edge technology, and so many unfamiliar platforms. Much more ambiguity, and insanely thought-provoking brainstorming sessions. 20 hours absolutely felt as if they happened 4 days ago. And we are indubitably satisfied with how far we have come.
What we learned
The best part is that we learned a good deal about each other. And ourselves too. We are 100% sure that this awareness is going a long way with us.
What's next for Panah Inc.
Scalability. We wish to maintain a centralized cloud space for all the secured translations of photos. We want to scale this to all the branches of major hospitals, and wish to include all renowned medical practitioners. We also wish to incorporate a feature with which, ordering food would be possible using the same app.
Log in or sign up for Devpost to join the conversation.