SOLVOID

Inspiration

we have seen our neighbours who are deaf and dumb and they talk in sign languages which other people are not able to understand so we think of creating this sign language detection model so as to make the people understand what these people wants to convey.

What it does

Our project is a transformative solution to empower individuals with disabilities. Leveraging advanced computer vision, machine learning, and deep learning, our real-time translator converts sign language gestures into written or spoken language. This real-time recognition system serves to bridge the communication gap between individuals with speech or hearing impairments and those without, offering an accessible means of communication and fostering inclusivity.

How we built it

first of all we have created our own dataset by using opencv and cvscan . then we have trained our model using ADAM optimiser and CNN . then we are predicting and displaying that particular letter of which the sign is given by the person in the camera frame itself.

Challenges we ran into

First we have trained model on raw photos captured by us but we couldn't see the positive result. then we discovered the reason behind it which was that it was also using the surrounding data for training. then we dug deeper found some results to detect hands and passed the the picture through many preprocessing steps before training then we finally came but we proudly submitting now

Accomplishments that we're proud of

we created the dataset and we have trained our model on the dataset that we have created .this project was our first venture of its kind and use successfully created the dataset and process it, trained it and made prediction and all that. in hours we able to complete this successfully is what we made proud of.

What we learned

we have learned about CNN and various other things like python libraries such as opencv and cvscan to detect hands. we have also learned how to train the model on machine learning etc.

What's next for SOLVOID

we will look forward to expand it to detect the words or the phrases used by the impaired people. we will also be expanding it to create the line or a sentence that a person wants to convey. we will be integrating it to the video calls as well.