**7. 3D reconstruction techniques for cultural heritage**

AR technologies have become increasingly popular. These techniques are not just practical for developers of AR system, but also to the scientific community. The standard approach to create a 3D model is to build it from scratch using tools such as the unity 3D programme, which provides building blocks in the form of primitive 3D shapes. Many new technologies aim to increase the level of automation and realism by beginning with the real images of the object or converting it to direct digitisation using a laser scanner [28].

#### **7.1 Image-based modelling**

This technology includes vastly available devices, so the same system can handle a wide range of objects and scenes. These systems have the ability to create a realistic model, and those rely on photogrammetry have high geometric accuracy. This technique is usually used for geometric surfaces of architecture objects or for modelling precise terrain. It uses a mathematical model to capture 3D object information from 2D image dimensions or obtain 3D data using methods such as shading, texture, theory, contour and 2D edge gradient [41]. Deriving 3D measurements from images naturally requires that interest points be appearance in the image. Often, this is not potential, either because the area is hidden or covered behind an object or surface or because there is no mark, edge or visible feature to extract [28]. The main goal of image-based reconstruction is the ability to represent arbitrary geometry. For modelling complete geometric structures, it is usually necessary to remove the labour-intensive task through this approach [41]. The mechanism can also deal with the real-world effects that images take, but difficult to reproduce with the customary graphics techniques.

### **7.2 Range-based modelling**

3D geometry information for an object can be captured directly by this technique [28]. The 3D measurement of images requires that interest points or edges be visible in the image, which is not constantly possible. Illumination or ambient light problems can impact the extraction process of such points and edges. Active sensors, for example, laser scanners have the ability to avert these restrictions by creating features on the surface using controlled light projection [41]. Many range sensors are produced organised points, in the form of an array or range image, appropriate for automatic modelling. However, texture information or colour can be attached from the scanner using colour channel or from separate digital camera [42, 43]. High-resolution colour textures that obtained from separate digital camera help to create of realistic 3D models. Generally, a single range image is insufficient to cover any object or structure [42]. The amount of necessary images rely on the shape of the object, the amount of self-locking and obstacles and the size of the object compared to the sensor range [41]. In order to wrap each aspect of the object, it is mostly required to perform multiple scans from various locations, which is commensurate to the size and shape of the object and occlusions. The alignment and groups of the various scans can affect the final accuracy of the 3D model, where each scanner has different range of resolution [28]. In addition, this technique can provide accurate and complete details with a high degree of automation for small and medium size objects, which reach the human size [42]. There are two major kinds of range sensors: triangular based and based on the principle of flight time [41]. Triangulation-based sensors are working dependent on project light in a known direction from a known position, as well as measure the direction of the returning light through its detected position. The measurement of accuracy depends on the triangle base relative to its height. Sensors based on the principle of flight time measured the delay between emitting and detecting reflected light on the surface, thus, accuracy does not quickly deteriorate as the range increases [44]. Time-of-flight sensors have the possibility to provide measurements in the kilometre range.

#### **7.3 Image-based rendering**

Image-based rendering used images as modelling and rendering primitives [28]. Image-based rendering uses images directly for creating new views for rendering without explicit geometrical representation. This technique is a significant mechanism for generating of virtual view, where certain objects and under particular camera motions and scene conditions. From the image input, this technique creates a new view of the 3D environment [41]. This technique has the feature of creating realistic virtual environments at speeds independent of scene complexity [42]. Image-based rendering depends on accurately knowing the camera positions to use automatic stereo matching, where the absence of geometry data, requires a major number of carefully spaced images to succeed [42]. Most of image-based rendering correspond to hybrids image-geometry, using means of the equal amount of geometry ranging from per-pixel depth to hundreds of polygons [45].
