Nucleus
All the great ideas begin with pen and paper. It's easy and accessible. But what's next? How can we fill in gaps between hand drawing and the digital world?
DoodlAR presents an easy way to convert your interior design idea into digital form both 2D and 3D.
Keep your memories where it belongs
Client: Confidential Client

Role: Experience Designer

Responsibility:
Product Design, UX/UI Design, User Research

Time:
2020 Mar
“There’s nothing like drawing a thing to make you really see it”
- Margaret Atwood
Context
Explanation of hacking artssadoifjasjdfklajkldsjlfaj

Idea starts from hand sketch still, there’s a gap between hand drawing and the digital world.
How can we fill in gaps between hand drawing and the digital world?
How can we convert hand drawing to a digital product or service?
In today’s vast e-commerce landscape, the product search experience lacks imagination.

When you are moving in, or need to change furniture arrangement, people think before they actually jump into the action.
Furniture arrangement is a time consuming, tiring, committing job. People use brief sketches in advance to solve this problem and reduce the effort as much as possible.  

While sketches are an easy and quick way to prototype furniture arrangement, there is often a gap in real life and sketches since sketches can only provide one perspective, most of the time 2D top view.

Furniture arrangement needs to be understood in context, from different perspectives.
Solution
Through AR experience, view actual furniture from Wayfair. Easy to move around furniture and simulate the best arrangement
Browse items and save items in carts in this app and also save the scene for later
Switch hand-drawn sketches into 3D furniture models using computer vision. - Doodlar gets smarter every time using machine learning
Technology
How can DoodlAr understand users' hand drawn sketches?
Machine Learning; CNN image classifier with ml.js
To interpret human drawing, the team used DoodleNet from ml.js and CNN image classifier to train. DoodlAr will get smarter as you use it more often and time goes by.
How can DoodlAr bring users' sketches into practical interior design simulation experiences?
Image target recognition Augmented Reality with Wayfair 3D models sets
To interpret human drawing, the team used DoodleNet from ml.js which is similar to Google Quick Draw API and CNN image classifier to train. Doodlar will get smarter as you use it more often and time goes by.
User Journey & UX Construct
Journey
UX Construct
App Design
AR Experiences - Detect sketches, view 3D models
AR Experiences - View Product information & Save scenes
Saved Scenes - View all your saved scenes & products, and export a floor plan that has been created automatically
Explore - Explore other people’s spaces out in the world and get inspired
Product Demo
Hackathon Winner and Next Steps
We had learned so much from such a short amount of time in the hackathon sprint. Here are some key action items we would like to proceed as moving forward: 

• Add VR mode and enhance social aspect of it
• Curated search results based on users' doodles
• Create educational contents around interior design
• Expand to entertaining applications