AR Furniture: Integrating Augmented Reality Technology to Enhance Interior Design using Marker and Markerless tracking

Purchasing products for interior design always has a problem that the purchased products may not satisfy customers because they cannot put them in their own place before buying. The purpose of this research is to study and develop an android application called 'AR Furniture' with the use of Augmented Reality technology for design and decoration that will help customers visualize how furniture pieces will look and fit (to scale) in their homes and also can provide details of products to support customer decision. This application is a prototype to find out factors affecting the design and tracking of AR applications. This paper presents three factors that are important for designing and tracking AR applications. The principle of the application is started with analyzing images from the rear camera of a smartphone or tablet using marker tracking technique for displaying product's details and markerless tracking technique for displaying 3D models, performing feature tracking, and calculating positions to display a 3D model over the real world image. The implementation of the application can be split into 2 parts: Part 1 Creating 3D Models using Autodesk 3Ds Max and Part 2 Developing the application using Unity3D and Kudan Augmented Reality SDK as an engine for image analysis, image processing and 3D model rendering. Then we performed three experiments to test the application, 1) Image analysis with marker tracking 2) Image analysis with markerless tracking and 3) User's satisfaction of using the application. The results show that image analysis with marker tracking works well using markers which their size should not be less than 200 x 200 pixels, the distance between the camera and the marker should not be far more than 60 cm. Image analysis with markerless tracking works well with surfaces having a lot of features and at light levels of 100--300 lux (indoor light levels) with 70% accuracy. The user experience evaluation shows that the weakness (2.86 out of 5 points) of the application is when a user found a problem in the application they would need time to solve it. The user experience evaluation shows that the strength (3.93 out of 5 points) of the application is the application can show 3D Object that meet user satisfaction. And the average overall user's satisfaction come up with 3.93 out of 5 point evaluation score. From the experiments, the application should be modified for better performance such as develop various maker patterns using QR code or barcode, distinguish walls and ceilings so that the application would show 3D objects on them properly, improve light robustness and make 3D models more realistic.