Our personal experiences with disabled relatives and friends brought to our attention that not all places accommodate the disadvantages of people with disabilities. Due to the lack of transparency regarding accessibility between such places and customers, many will visit tourist attractions and find that it is not accessible for them and leave, dejected. We hope to ease the burden that some people with disabilities face and encourage companies and organizations to improve accessibility.

What it does

Users can upload tourist attractions and rate them. Other users can view the ratings on different metrics. There is also a comment function. We have a face recognition ml model that recognizes features from an uploaded picture so that visually impaired people can understand what is happening in pictures. It then reads it aloud with text to speech. We also have speech to text technology.

How we built it

We used many different platforms and languages for this, which was one of our main issues. We used pycharm for the python and html files. We used flask to link python and html and we used sqlite as a database. We tried using Tensorflow for our ML models, but that did not work, so we ended up using Colab. We also implemented apis from face++ and google speech recognition among others.

Challenges we ran into

We faced multiple obstacles during this 50 hours work period including issues with Tensorflow and Flask which we were not able to install on our computers (with the exception of one person). Furthermore, we struggled with integrating our Java ML with Python features and our database website. Additionally we had issues with our ML libraries and communicating to API services

Accomplishments that we're proud of

We’re proud of our ML implementations because none of us had ever worked on ML models before. Also, this was two out of our three members' first hackathon. Also, it was very difficult to implement the API, and we were very proud of that. Also, we almost set up tensor flow after spending five hours on it, so we were proud of that. Finally, we were glad about including so many working features.

What we learned

We learnt about how to use Python for Web and how to use SQLITE. We also learned to use flasks and incorporate it into our code. We learned how to outline and form ML methods. We learned how to make full stack code and integrate our different features from multiple languages onto one website

What's next for Helping Hands

We plan to fully integrate all our ML with our database and also deploy a domain for our website. In addition to this, we want to format face identification data better and make it more organized.


(sorry the faces aren't there, we recorded it with video, but it didn’t show up)

Share this project: