The Value of Existing 3D Data in the Age of Augmented & Virtual Reality

Stefan Minack

March 14th, 2017

In this article, we take a look at what is necessary and what you can achieve with the data you might already possess (i.e. OBJ, FBX, Collada) to make it work in the context of virtual and augmented reality. As AR/VR technologies are being considered to change the way human beings interact (as was the case with the advent of the PC, Internet, and mobile), applying these technologies in practice will surely affect the way we convey messages to our customers, the way we do business together, and the day-to-day communication with our co-workers. These technologies equip us with visualization, information, and playfulness at the same time, and they are ready at hand wherever we go.

Getting Started

Working with real-time visualization can be rough. It can be difficult because using Virtual and Augmented Reality is no less than rendering images in real time, but with plenty of sensor data or processing images from the device’s camera on top of it. Luckily there is enough software and hardware around that takes away the pain of dealing with the second part of that equation: processing the sensor data. Software tools such as Vuforia, Wikitude and also hardware and software embedded into devices such as mobile phones and tablets.

Nonetheless, we should keep in mind that processing sensor data still uses some of the system's resources and limits the space and computation cycles we will have left to render the visualization. We do have some constraints we need to optimize the content for, to run as smoothly as possible. This means we need to take care of things like the limited memory and processing power of the device, something that is very common in designing games. So for our use case, the 3D data can come in plenty of different formats and flavors. To make it feasible for our application we need to do some work.

Two Types of Data

First, there is data that is only created to look as good as possible but does not have to be accurate in any kind of way. It is amere visually correct representation of an object. If you are working in the field of creating high-end CGIs, it normally does not matter if an image takes two or two and a half hours to compute. It also doesn’t matter if it takes 30 or 40 gigabytes of RAM to draw those images to a hard disk. Working with animations makes the processing time decrease, but we are still far away from creating those images in real time.

Secondly, there is the kind of data that is created by engineers—data that is a virtual representation of an actual physical object, with all its mechanical properties. This can go down to the point where every nut and every bolt on a machine, every small detail, is present in the model. Most of the time, this kind of data is problematic in different kind of ways. On the one hand it is not possible to display it on, let’s say, a mobile device. On the other hand this level of detail is usually confidential data, something an engineer does not want to have floating around somewhere on the Internet.

In both cases there has to be some kind of data reduction. This could be achieved by just doing manual labor, in terms of reducing objects that are not necessarily needed for the presentation. It could also be achieved by recreating models with the right amount of details, or even by using algorithms, and by recreating or mimicking the visual characteristics of the materials used on the object we want to present.

Conclusion and Outlook

We are still talking about data that is already there. Data that was already created in the process of designing and engineering a product, data we only have to take care of in the right and feasible way. While this real-time representation of a product cannot compete with the visual fidelity of a pre-rendered image or video, it can carry a lot of value by adding different levels of interaction and information to the 3D model.

Example of a 3D tour with interaction (Vuframe)

Practically speaking, it can be a deal breaker if a consumer sees that an object (a couch, a TV, or a piano) will fit into her home or not. AR helps her to bring immovable goods into her home. With VR she could do it the other way around and explore spaces that are either not existing yet or hard to reach (future properties or production centers on the other side of the globe). In short, we are all visual thinkers to a great extent and AR/VR will become the leading technologies facilitating this aspect.

Sofa in life-size Augmented Reality based on CAD data (Vuframe)

For companies and freelancing professionals that are into creating or working with 3D data, this is a huge step. They can utilize the data they have already created, optimize it, and put it into a better context by adding interaction, different layers of information that can be displayed, and use it for an example in sales and marketing activities, or even employee education. And by keeping in mind that this is an option, it is possible to create the data in a way that they can skip most of the optimization process, which reduces costs and production time.

About the Author

Stefan Minack is a 3D artist and head of content at Vuframe. He’s been working in 3D content creation on a daily basis for the past 8 years. Vuframe’s mission is to democratize AR & VR by removing the tech barrier for everyone.