Synthesizing reality for realistic physical behavior of virtual objects in augmented reality applications for smart-phones

Techniques in Augmented Reality

“…It helps in the localization of 2D features that are extracted from images and are compared with known 3D models. [11] represents a framework for augmented reality applications that run on smart mobile phones and enables realistic physical behavior of the virtual objects in the real world. It also focuses on the reconstruction of the 3D representation of the real world.…”

“…2D views of the 3D model are used to identify potential locations of the objects to be searched. These 2D views consist of two types of edges they are Essential edges [11] and Optional edges [11]. Using graph-matching algorithms, potential locations of the 2D views inside the 2D image graph are evaluated.…”

“…2) 3D Vision: In the 3D vision part, the 2D features are compared with CAD data, containing highly visible edges, faces and texture information [11]. With correspondences of image features and 3D-model features, hypotheses for the orientation of the model relative to the camera are generated.…”

“…With three 3D-2D-edge-pairs it is possible to compute the relative orientation of the camera to the object, therefore, hypotheses consisting of three edge-pairs are used. Fig.3 Edge Parameters of given object [11] In object recognition systems, depth data provides otherwise absent geometric data, a potentially powerful tool for discrimination between objects [3]. Depth images contain local information about object geometry but also provide information on global geometry, object position, and object shape.…”

“…There are three different slice data type: i) Grey Image Line: This data type stores the lightness of the pixels of exactly one line of the image. ii) Gradient Line: This data type stores the gradients of the pixel as a result of the Sobel filter [11]. It also stores the data for one image line.…”