Visionary

• 

  Let's make this world more accessible

• 

  Training AI model

Inspiration

We are inspired to bring a change in the life of visually impaired people.

What it does

It takes in real time motion through webcam and describe the surroundings so that visually impaired people have less problems in their day to day life.

How we built it

We used pretrained YOLO's v8 in python and trained it with day to day objects' images. Based on the image we got the type of object like car, person etc. and then the model speaks out what the object is.

Challenges we ran into

Training AI model was one of the biggest task we had to accomplish, but apart from that there were many problems in audio outputs and object detection from live feed.

Accomplishments that we're proud of

Training an AI model and deploying it successfully.

What we learned

Connecting backend to frontend, training AI model.

What's next for Visionary

We want to improve our skillset so that we can make this a fully functional app as originally intended.

Built With

• firebase
• kagel
• python
• react
• tailwind