Augmented Reality has begun to get more widely noticed and for good reason. It’s potential as an advertising tool is almost limitless. The numbers of applications that have been developed and are out there for the public, like the ones listed earlier, make have got people interested in the technology. It is making our life easier already and is allowing us to do things that, only 60 years ago, were considered Science Fiction. Part science fiction, part a reaction to today’s increasingly overwhelming constant barrage of digital content, one thing is for certain, though Augmented reality is an important step on the road to making technology more understandable and useful. This project aims at implementing augmented reality with the help of a real wall. The wall holds the virtual images and videos projected as 3D photo frames over multiple markers. These multiple markers are detected by the system simultaneously and the virtual objects corresponding to each of them are augmented upon them in the scene.
Requirements for Augmented Reality in larger 3D space are
- High Quality 3D Rendering at Real Time which can scale from Smartphone to Caves
- Multimodal Interaction with Multi Touch Systems and Haptic Interactions
- Development of VR Systems which include Tiled Display, Touch Screen & Caves and Movable Screen
- Computer Vision (CV) based Tracking using Markerless Tracking with CV and Sensors
Augmented Reality Framework is as shown below. In this framework, Content and Player is separated. The design should be such a way that it should be Platform independent. For communicating with other systems, interoperability and smooth integration should be offered.
AR enhance user perception of interaction with the real world. The information conveyed by the virtual objects helps a user perform real world tasks. The virtual objects display information that the user cannot directly detect with his own senses. AR is a specific example of what is known as Intelligence Amplification (IA), using computer as a tool to make a task easier for a human to perform. Even though markerless AR systems are able to perform very well in certain settings and compute the pose on the planar surfaces without any user interaction, the necessity of abundant features on the scene makes it unsuitable for our application, which basically tries to augment paintings on a background, which is most of the time a plain wall. Other than this, we would require application to run almost in real time on a handheld device, which mandates the use of a computationally simple method. Marker based systems simply provide good results under these two constraints. First of all, they can locate the marker by using simple techniques that don’t require much computation. Moreover, they can almost always give reliable results when the marker is in the viewpoint of the camera.