A Marker-based AR System on Image Shadowing for Tourists

Abstract

At tourist attractions, in more cases, the human guides charge hefty amounts, and a manual is only available in some places. This work is about building a traveler-friendly application that can introduce AR-based experiences. AR is an immensely high technology, but it still has defects that are stopping the community from implementing it on a large scale. Nevertheless, it can change the way we look at the world. Marker-based AR provides a very immersive experience and closely mimics the real world; with this, user engagement with the user becomes easy. This technology demands high computational power and captures a sizable amount of memory. Android Studio was used to make the program, and we used the ARCore library as the AR rendering engine for our app. In our app, when we project the phone at an image, a video (AR scene) related to the image will be played, replacing the image in our application. A cloud database and internal storage store the augmented scenes and the corresponding images. Marker-based AR technology and image shadowing techniques achieve this. We used the historical image or scene as the anchor point marker. Designed an AR model that takes each frame from the camera feed, processes it, and finally renders the AR objects that will be projected into the scene. We tackled Augmented reality problems like light estimation, motion tracking, environmental understanding, and by fine-tuning the application. Through this implementation, people will be able to use it in tourist areas, and it has the potential to be better than a human guide.