As people grow accustomed to effortless outdoor navigation, there is a rising demand for similar possibilities indoors as well. Unfortunately, indoor localization, being one of the requirements for navigation, continues to be a problem without a clear solution. In this article, we are proposing a method for an indoor positioning system using a single image. This is made possible using a small preprocessed database of images with known control points as the only preprocessing needed. Using feature detection with the SIFT (Scale Invariant Feature Transform) algorithm, we can look through the database and find an image that is the most similar to the image taken by a user. Such a pair of images is then used to find coordinates of a database of images using the PnP problem. Furthermore, projection and essential matrices are determined to calculate the user image localization—determining the position of the user in the indoor environment. The benefits of this approach lie in the single image being the only input from a user and the lack of requirements for new onsite infrastructure. Thus, our approach enables a more straightforward realization for building management.

No abstract

Looking at the concept of the virtual model of Terezín Memorial, the model consists of two main parts: the geometric 3D model which depicts the place and the lexical base of data which describes the history of Terezín, mainly during the II. World War.Taking a closer look, the heterogeneous nature of existing both analogue and digital documents about the history leads to use a content management system (CMS). CMS is used, together with a relational database, for the lexical data. Each document in CMS has its unique identifier, identifier of a place to which is referred to and two dates referring to a time period. The spatial part of the Terezín model consists of detailed models of all historically valuable buildings and constructions. These models are complemented with less detailed models of the rest of buildings and constructions in the Terezín town.Both lexical and spatial part of the virtual model is filled up with large amount of data. Therefore it is crucial to build such a method of (both lexical and spatial) data selection, which is fast and serve relevant information to the user. Whereas well known one dimensional data indexes can be used for descriptive data, situation is a bit more complex in 3D, where common geographic (two dimensional) data indexes cannot be used. In 3D, size of bounding box (BB) of each potentially portrayed object is calculated, based on the observer position, its view direction and the distance from the object. When the size of BB of each object is calculated, different level of detail (LOD) of each object can be displayed. Different LODs are used also in two dimensional maps, where they are used for creation of a scale dependent map, but there is again principal difference between 2D and 3D. While in 2D always just one LOD at a time is portrayed in the map, objects closer to the observer are displayed in higher detail then farther ones in 3D. It leads to a situation, where objects are displayed in different LODs in one 3D scene. This issue poses a major challenge to a creation of a multi-scale 3D model, because different LODs have to share major shape (at least footprints and heights). Moreover the virtual model has to be also time aware, therefore each piece of information in the database has to have a time period for which it is valid.

As people grow a custom to effortless outdoor navigation there is a rising demand for similar possibility indoors as well. Unfortunately, indoor localization, being one of the necessary requirements for navigation, continues to be problem without a clear solution. In this article we are proposing a method for an indoor positioning system using a single image. This is made possible using small preprocessed database of images with known control points as the only preprocessing needed. Using feature detection with SIFT algorithm we can look through the database and find image which is the most similar to the image taken by user. Pair of images is then used to find coordinates of database image using PnP problem. Furthermore, projection and essential matrices are determined allowing for the user image localization ~ determining the position of the user in indoor environment. Benefits of this approach lies in the single image being the only input from user and no requirements for new onsite infrastructure and thus enables a simpler realization for the building management.

No abstract