Photogrammetry is an excellent tool to use in eLearning. It allows us to bring accurate representations of objects or locations to an unlimited audience. Before exploring how photogrammetry can be applied to eLearning, let’s take a look at what exactly it is:
In the above illustration from http://schnellinformatics.com/photogrammetry.html, we can see how photogrammetry works. If you take a single photo, the resulting image is 2-dimensional and the computer has no way of knowing which objects are near and which are far. A human observer will be able to make this distinction based on their understanding and recognition of different objects in the photograph. If you, however, take multiple photos from various angles of the same object, a computer will be able to compute distances based on triangulation of similar features in the images. Using trigonometry, the software calculates the position in 3D space of camera locations and angles, and the position of many matching points from the photographs.
The way photogrammetry software works is by comparing photographs of the same object from different angles. It finds points in the photograph which match, and uses these as points of reference to calculate locations and angles of the object and cameras. The more photographs you provide, the higher the accuracy of triangulation. The photogrammetry software creates what’s called a point cloud using all the matching points on the photographs. The point cloud is essentially a series of dots, placed correctly in 3D space based on calculations from the photos. When you have enough of such points – hundreds, thousands, sometimes millions – the objects in the photographs will start to take shape in 3D space.
The above image shows the point cloud of an Austrian church in Graz. The image and the software used to create it can be found at http://www.geoconnexion.com/news/microsoft-announces-release-of-ultramap-v30/. You may notice that there are “holes”, or incomplete sections in the point cloud. These represent areas not covered in the photographs provided to do the calculations. For a complete scan, you must provide photographs from many angles, covering all the surfaces of the object you wish to scan.
After the point cloud is created, the software will create a 3D mesh of your scene or object for use in other 3D modelling software. Depending on the software you use, there will be options as to how dense or accurate you want the 3D mesh to be. Generally, this mesh will have a very high poly count to represent the scanned object accurately. This would be a problem if you wanted to, for example, use the object or scene in a game where poly count must be limited. There is a solution, however! You can bake the high poly mesh to a normal map, and apply it to a low poly version, preserving all the beautiful detail. If you want to know how, just search on the internet – that topic is beyond the scope of this humble article.
The above image from http://www.ysma.gr/en/geometric-documentation shows the point cloud (left) and 3D mesh (right) of a historical statue from Greece.
Let us return to the question of how we can apply photogrammetry to eLearning. When it comes to historical objects or locations, we cannot always bring our viewer to the place – sometimes we have to bring the place to our viewer. Being able to recreate objects and places accurately in 3D, we can create a simulated environment where the viewer can explore the location or the object almost as if they were there. This can also be applied to medicine, for example, where the student is able to view and manipulate organs scanned from actual samples. Extremely accurate scans can be made using laser scanners rather than conventional cameras. Other applications of photogrammetry reach any number of sciences, from history to geography, chemistry and physics, botany, mechanical engineering, and so on. If there is an object or a location that needs to be represented accurately in 3D, photogrammetry may be the best way to do it.
Photogrammetry can be used in gaming as well, to create absolutely stunning visual scenes. One such game that used this tool to create a beautiful and immersive world is “The Vanishing of Ethan Carter”, where photogrammetry is used to create scenes from actual world objects. The difference is visible:
If you would like to learn more about how photogrammetry can be used in eLearning, or to see how our company approaches eLearning in a new and immersive way, please visit our website:
We specialize in bringing eLearning to the next generation through a variety of highly effective means, from traditional classroom training to eLearning and eGaming designed to maximize learning effectiveness and retention.