Photo Connect

Facial recognition algorithm
UI Concept
The photo gallery UI interface
Add friend UI interface
UI product

Inspiration

Group photos are incredibly fun, especially when there are huge groups of people in it. However, frustration arises in the after-phase when the sole person who has the image on their phone is asked to share it with everyone that was present in the image. This person has to scroll through their collection of contacts to send the image to every contact. This is annoying when there are 30+ people in the picture and they may make the mistake of missing one or two people.

What it does

PhotoConnect is like any other social media app. Users are able to make a personal account, expand their network of friends, and send images to people in their friend group. Users first can upload the image to their account to be stored on the cloud database associated to them. Then users can press a button that runs a facial recognition to detect the set of friends the picture needs to be sent to. Once the user hits send, the images are sent to the set of friends recognized.

How we built it

Our full-stack project utilizes React Native for designing the frontend UI experience as well as the backend processing necessary for our app. To support the backend, we used FireBase for login authentication and storing data for each user (e.g. username, email, password, friends, images) on a cloud-based storage system. Finally, our facial recognition algorithm is written in Python with the help of the DeepFace library.

Challenges we ran into

React Native and FireBase were fresh new tools for most of us. Setting up the packages necessary for the tools was one challenge, as well as finding the right functions to be used in our codebase.

Accomplishments that we're proud of

We're proud of our project idea and being able to design a facial recognition algorithm to accomplish the task. Even though authentication isn't the main part of our project, we are also proud of being able to build a relevant feature in today's society, making our login process reliable and secure. Using a cloud database is also an interesting and proud experience for all of us.

What we learned

We earned valuable experience in full-stack development as well as the application of new tools such as React Native and Firebase.

What's next for Photo Connect

We hope to strengthen our facial recognition and friend-network logic by providing ways to detect mutual friends in images. We also hope to enhance the user experience by expanding the set of features given by our set of screens. We hope to have "In" and "Out" screens that display the images being sent to the user and the images sent by the user, allowing the user to accept and decline images.