-
-
Welcome
-
You can take a photo on your question, or upload your photo. You could also get the search history.
-
You can get the answer after a short time. Click the Best answer, you can get more information on this answer, we also could more info.
-
This is the rough result from our server.
-
This is the photo we used in test.
-
This is the course links you will get from us and the link from Chegg
Inspiration
We got the idea from Chegg and Coursera. We think the refugees really need to get tutor and teachers. But the resources of education are not enough to share with everyone. On the other hand, we really want to help the disabled women to get the tutor.
When we have some questions in our assignments. The first thing jumps up from our mind is using the search engines like Google or Yahoo to find out how to do it. But it is really hard to the disabled women, because type in the questions is not very convenient, also when you want to type in some math questions. The symbols from math are the trouble when you want to type it.
Then we want to develop an App which you only need to take a photo on your question, and then, the answer will show up!
What it does
First, you take a photo on your question. It will upload the photo to our server automatically.
Then, our algorithm will get the words in the photo then search it on Chegg, Yahoo Answer, and Coursera to help you get the question's answer.
Finally, the answer is on your phone, very easy to get, but no need to type in any words.
How we built it
We used the OCR technology from Google Cloud Platform API to get the words from the photos. And we also tried to use Google's TensorFlow open source platform try to training our server to get the more accurate answer from the photos.
Challenges we ran into
Time! The time is very limit. Our project is pretty complicated. So did not finish everything, especially the deep learning part.
Accomplishments that we're proud of
Now, your can take the photo to get the words from the photo, and you can find the answer.
What we learned
How to use the TensorFlow platform and Google cloud platform. Also, we learned how to get the answer from the search engine result.
What's next for SnapKeys
We want to finish it as a perfect project and try to get the API from Chegg and Coursehero. Now it is only could get recognize the English questions, we hope after we trained it more times. We can get more accurate result. It needs more data from Chegg or Quizlet.
Log in or sign up for Devpost to join the conversation.