ASL Signup/Signin Frontend and Backend Project

[TayKimmy]

Goal

The overall objective of the ASL at this time is to integrate ASL into the frontend and allow users to create a video of their sign and login with their sign. All the data has to be added to the backend and associated with the Person object so that frontend and backend work cohesively.

What will be shown to teacher and audience

A functional login and signup page where the audience can create a new user with their own unique sign. They can then login with the user they just created. --> maybe show key frontend/backend code + AI model that works with ASL

Frontend Design

Login:

Signup:

Frontend Progress

• Take a 10 second video of the user's sign when they create a new user
  • Send this video to the backend and convert to MNIST data
• When a user logs in, compare the sign currently on the screen with the sign associated with the user
• Let user in if correct
• Make sure cookie is not created after initial login (username and password) so that users can't go directly to the profile without doing the 2nd login
• Delete cookies if ASL login is wrong

Backend Progress

• Update the Person repository so that it includes the data - maybe
  • Me and Anthony were discussing what exactly needed to be added on to Person, and we thought maybe we can just use their name. The MNIST data has a column with ID, which associates to different names. We can access their data through their name.
• If we do go in this direction with using their name, we have to update the part where we define their names for the MNIST data. Each time a user is added, we have to add their name to that list.
• Right now, our backend runs manually, we have to get the model and data to work each time there is a frontend request.
  • This means fixing and adjusting backend.

Schedule

Monday

• Work on recording a 10 second video in signup
• Get the individual frame in login

Tuesday

• Send the data to the backend and update the backend to add a new Person
• Train this data using the model and add to MNIST data

Wednesday

• Create fetch methods in the Frontend and try to compare the sign from the frontend with the backend

Thursday

• Work on cookies

Friday

• TEST TEST TEST