Imagine providing 3-D (three-dimensional) models for learners to study—medical models, 3-D reproductions of sculptures, geographic locations, even of people. Now imagine creating these models easily and inexpensively. Using a 3-D model moves learners toward fully immersive or augmented reality experiences and adds exponentially to the potential for interactive learning.
Creating a 3-D model using photogrammetry is a shoestring method of creating a virtual object for study. Learners can view and interact with a 3-D model using their computers or mobile devices, turning the model around and looking at it from various angles. Online clothing retailers, for example, sometimes use this technology to show the fit of the clothing from all angles on a 3-D model that stands in for the customer.
Photogrammetry and smartphones
A 3-D model is produced from digital still photographs, not from video. An eLearning developer can create some 3-D models—of a person, a sculpture, or a building, for example—using a smartphone camera and free or inexpensive software, such as the 123D Catch app. Creating 3-D models of larger objects or locations or generating a digital elevation model of a tract of land requires aerial photography.
The process of photogrammetry converts multiple overlapping photo images into a 3-D model of an area or an object. It’s also used to create maps and measurements.
- Aerial photogrammetry uses aerial photographs, taken from a drone or airplane, to create maps or digital elevation models. The camera shoots multiple overlapping images of the ground along the flight path.
- Close-range photogrammetry uses photos taken with a handheld or tripod-mounted camera; the results are converted to 3-D models using point clouds. A 3-D model of a building or an object of any size can be created this way; some models can be printed using a 3-D printer.
Though the technology is easily available and affordable, getting good results requires some preparation. Sketchfab, a website devoted to 3-D content, offers a wealth of advice and examples, including these basic tips:
- Any camera that takes high-resolution images can take suitable images for photogrammetry; fixed-length lenses will produce better results than wide-angle or other distorting lenses. All photos should be taken using the same focal length—no zooming.
- A tripod is invaluable to minimize movement and enable longer exposure times; any movement of the camera or subject can cause blurred, and therefore unusable, images.
- The photogrammetry software will produce the best results with textured items, such as rocks or tree trunks; shiny or very flat objects will not work as well.
- Good lighting is important, though it does not have to be bright. Even lighting will produce better images than subjects shot in shadows or changing light. If shooting outdoors, an overcast day is better than a sunny day with reflections and shadows.
Successive photos should overlap considerably. This enables the photogrammetry software to identify the same point in different photos and create an accurate 3-D model. To ensure accuracy, Sketchfab suggests roughly 60 percent overlap between photos. The number of photos needed will vary, but it’s hard to overdo it. Blurry photos must be discarded, and any extraneous photos can also be discarded; having too few images will harm the quality and accuracy of the model.
Photogrammetry software uses a series of triangulation algorithms to calculate the angles and distances between points in the images. It creates a “point cloud,” a series of dots that, eventually, forms a 3-D image of the object in the photos. If a section is missing or shadowed too darkly, it will not show up in the point cloud, and there will be a gap in the model.
Using photogrammetry projects in eLearning
Instructional designers can use 3-D models created using the photogrammetry process to enhance or create virtual environments for VR projects, such as simulations or games, or in eLearning that emphasizes study of an object or place.
The models offer realistic imagery for learners who are studying anything from anatomy to geography. While learners might not be able to see a work of art or an archaeological site in person or get to a geographical location, they can explore or examine it virtually, using a 3-D model. Medical students, for example, frequently study 3-D models of organs and systems; the models, created from actual samples, provide accurate representations of different conditions. While more complicated to create than models of small objects, a 3-D map of a location can be used to create realistic simulations or enhance virtual environments.
An eLearning module can use a 3-D model to test learners’ ability to identify parts of an object or to solicit suggestions for improvements to the design of a component; a 3-D map or elevation model can be used in eLearning to teach employees—park rangers or smoke jumpers, for example—the features of terrain where they will perform their work. Practicing and honing skills using 3-D models can be a cost-effective way to engage learners in realistic training and simulation exercises and improve overall performance.