Pre-construction Developments in Toronto | Example
2D Floorplan | Example
OpenCV x Python 2D to 3D Converter | Step 1
OpenCV x Python 2D to 3D Converter | Step 2
OpenCV x Python 2D to 3D Converter | Step 3
OpenCV x Python 2D to 3D Converter | Step 3
OpenCV x Python 2D to 3D Converter | Step 4
OpenCV x Python 2D to 3D Converter + Integrating to Unity3D | Step 5
OpenCV x Python 2D to 3D Converter + Integrating to Unity3D | VR READY!
OpenCV x Python 2D to 3D Converter + Integrating to Unity3D | AR READY!
Aside from my passion for programming, I always liked the Real Estate industry to the point 2 years ago, I decided to get my Real Estate License and in the meantime work with my mentor who has been specializing in pre-construction developments for almost 16years. Since then I learned a ton about the ins and outs of the industry that opened my eyes to many unnecessary and inefficient barriers in the buying process that hinder both the buyer and the developer.
What it does
My goal with this project is to shed the stigma of pre-construction projects and be a step closer to bridging the gap and provide a much painless and straight-forward experience in the process for both the buyer and developers by streamlining it. Introducing REALX8, the go-to platform for pre-construction/exisiting developments both residential/commercial. A platform for developers to easily market their projects at a much lower cost ( instead of spending 50k on printing flyers , the 2$/click google-ads fee or paying 4% commission to realtors who sell their units which usually costs 20k-50k ) and extremely easy process for potential buyers to find their desired property ( instead of signing up on some random realtor’s newsletter, and being bombarded with non-stop cold-calls all because you wanted to know the price range ).
Although this streamlining, solves a lot of unnecessary browsing/researching but it’s not enough to get rid of that uncertainty the potential buyer feels. And that’s where REALX8 shines by introducing one of its killer features which is giving the buyer a photorealistic simulation experience of the property they are interested in, by allowing them to walk around the unit, configure the look/color of the unit to their liking to get a clear picture on how the unit will look 2,3 yrs down the road when the project is completed. This feature solves a lot of the uncertainties a potential buyer feel but to be honest you can only make it realistic so much using a phone.
In order to go the extra mile and fully bridge the gap, the same simulation can be used using VR glasses such as Oculus Quest/Rift which can be stationed in the developer’s showroom/sales office to take a look at a much more realistic graphics using Ray Tracing technology for example which is only available in PC atm P.S which I still can’t believe how much progress we’ve made over the past decade, it’s mind-blowing.
As Jordan Belfods mentioned when it comes to having a smooth sales process, the salesperson, in this case, the developer's rep or realtors has to categorize buyers in terms of their level of interest. This will be the most accurate and photorealistic representation of the unit which is not only configurable like the phone version but also has the ability to add furniture to the property to make it even more accurate and make it look just they want it will be when they receive the property, and also what really happens in the background in terms of the science of sales, is that by having the buyer to come to the showroom and spending time decorating their property. You can detect their levels of interest and categorize the buyers in terms of how serious they are. This method provides a much more efficient process for both the seller and buyer.
Even though these features help the consumer a lot in terms of pre-construction development, we also developed a feature where a seller of an already existing property can draw/provide a 2D sketch of their property and using our own AI algorithm we can detect the shape/outline of the given drawing/floorplan and create the 3D version of it inside Unity3D and this can be accessible to other potential buyers of the property through our app (feature coming soon!). This is a great alternative to expensive cameras/services such as Metterport because not only it's free but also the potential buyer has the option to walk around the house using either their phone or their own VR headsets all because of the fact that the 3D object is built in Unity3D environment.
How I built it
For developing the REALX8 cross-platform app we used Flutter and for fetching our data we used Firebase and on top of that, we integrated the 3D graphics/objects and AR/VR features using Unity3D. For developing the AI algorithm we used tools such as OpenCV and Python.
Challenges I ran into
One of the most challenging aspects of this project was coming with the AI algorithm for detecting 2D things and converting them to 3D objects automatically, all by providing the picture of the designated 2D thing.
Accomplishments that I'm proud of
We're extremely proud of what we have developed within such a short amount of time and the skills we learned this weekend are definitely going the pave the pathway of our future.
What I learned
By developing our AI algorithm I now have a much better understanding of how Machine Learning/AI projects are implemented and now see the potential benefits it has for us in the future.
What's next for Fynd - Streamlining the Real Estate Industry | Flutter, Unity
We are definitely considering, polishing our app to hopefully see if we can not only pull this off in the real world but also be the change-maker on real estate pricing.
-This project was initially called FYND