Traditional 3D models made up of surfaces have for a long time aided us in visualizing buildings and spaces, but they often come at a cost: large models require a lot of storage and processing power, and can become incredibly complex to the point where they are difficult to navigate. As a part of our Selected by Sketchfab series, Sketchfab has their eye on a more efficient, increasingly common method of capturing architectural spaces; namely, point clouds. Point clouds are made up of a set of points located in a three-dimensional coordinate system, that when put together merely give an impression of the surface of an object, or the façade of a building.
The method is fairly simple. The collection of data points is generated by a 3D scanner that rotates while emitting a laser that measures the distance to points on surrounding surfaces. This data can then be converted into a polygonal model that can be rendered like any other 3D model. However, the advantages of keeping the scan in point form are what makes it great; the file sizes are much smaller, and the porosity of the point clouds make it possible to see through walls and surfaces, accessing "hidden" spaces and uncommon views of seemingly familiar surroundings. Read on to find out more about the possibilities and advantages that come with point cloud modelling.
1. Generating Large-Scale Overviews
Getting an idea of the scale and relationships between spaces within a site can be challenging and time consuming using traditional polygon surface modelling, and will certainly produce a large file to work with. Using the point cloud technique minimizes these problems, as can be seen with the model below:
Of course the resolution of a point cloud can be adjusted to include far more details, as in the following models:
2. Accessing "Hidden" Spaces
Due to the low density of points in the clouds, it’s possible to see through what would ordinarily be solid surfaces, as in the case of the passages within the thick stone wall in the model below:
The possibilities that come with this simultaneous visualization of multiple layers of space could be a fantastic tool for architects to fully grasp the locations they’re working with:
3. Experiencing a Space Through Virtual Reality
Scanning an entire building can consequently be paired with a virtual reality headset in order to re-experience, or share an experience of a space with others who may not have had the opportunity to visit the site. While this is also true of other scans with surfaces included, the same advantages of small file size and partial transparency mentioned above come into their own in virtual reality too, providing a smooth viewing experience and a sense of what's just around the corner, as demonstrated here:
The same can be applied to scanning landscapes, in order to experience the visual results of different typology:
When it comes to virtual reality, point cloud scans may be most useful for places with complex and detailed material qualities, where even if the scan is relatively small, the file size of a fully detailed model would be entirely unmanageable. This is the case in the following scan, which gives an excellent impression of the cabin's rough wood and overgrown roof without having to render the millions of faces that would be required:
4. Visualizing Simplified Geometry
Point clouds allow for the saving of models with complex geometry, without sacrificing enormous amounts of space and computing power, by minimizing the points collected in the cloud. Despite the perforated effect, the varied surfaces are easy to detect and view:
Here is a model of an interior space, simplified even further due to the dominance of flat surfaces. The contours generated by simplification can sometimes even be more beautiful than the original complexity:
