Inspiration
We like computer vision so we think that project would be the best for us to learn new ways of mining data from scratch.
What it does
The user provides the url of the house he or she wants to rate and then the app shows the pictures with the different characteristics, features and images stracted via the API and the webscraping. We also provide the possibility to the user to do age recognition using the camera and comparing that age to an age we approximate the house will fit in (thank's to the information extracted previously). Thank's for this we can estimate if the house will match with the client
How we built it
We had used the restb.ai API to identify and quantificate different house features and characteristics. To obtain the pictures and the data we have done web scrapping using requests and selenium on the fotocasa webside.
Challenges we ran into
We didn't know anything about computer vision, web scrapping and scripting. So for us the whole project is full-time challenge (we didn't sleep for 48 hours)
Accomplishments that we're proud of
We learn a lot of new ways to extract data from several sources and use it to provide some "predicitions"
What we learned
A lot of things, but the most important point is how to work in pairs
What's next for HouSee
We would like to add new pages to extract photos and information
Log in or sign up for Devpost to join the conversation.