With the developments happening in the realm of software and hardware breaking new ground daily, it is not far that our reality can be manipulated to suit our everyday lives and enhance our evolution. Welcome to a world of new reality. Augmented Reality manipulates existing dimension giving us a better world view. The rapid growth of mobile technology and ease of application development processes has given the push  to develop Augmented Reality applications at a large scale. Today there are many AR apps that solve real problems like measuring distance in real time, interactive learning and gaming. This read will talk about AR applications and it's attributes like techniques, user-interactions and their outcomes in detail.

Table of contents

  1. Understanding AR
  2. AR Tools and Techniques
  3. Features of AR core
  4. Studying the Spectra
  5. AR App Insights Visualization
  6. Inferences

Understanding AR

In simple words AR is an interactive, reality-based display environment that makes use of computer generated display, sound, text and effects to enhance the user's real-world experience. It is the integration of digital information with the user's environment in real time. Augmented reality builds on top of the existing environment and overlays new information unlike virtual reality which creates a completely artificial environment to immerse the user.

AR Tools and Techniques

AR today is being explored across fields using either applications or integrated software and hardware systems to solve real world problems. They are classified based on the purpose, use case, and application. The technology that exist are Marker based and Markerless. Marker based technology makes use of images present in the scene in order to effect overlays. Markerless technology uses geo tags to decide placement of objects or overlays.

Vuforia is a widely used SDK to develop AR applications for mobile users. It uses Computer Vision techniques to recognize and track planar images (Image Targets) and simple 3D objects in real-time. The target images should be uploaded and verified in the Vuforia Target Manager site to generate target image database, which is used in the AR application.  This image registration capability enables developers to manipulate virtual objects and other media, referencing real world images when seen through a mobile camera.  Other familiar tools were Meta IO, ARToolKit, AR SDK, Wikitude, Layar SDK.

With AR now showing potential as a software solution across industries the tech industry has shown keen interest in adopting the technology and building on it. Tech giants like Apple and Google are pumping in their resources to offer AR as a feature with their devices.

The introduction of ARCore and ARKit has now changed the game. Both the SDK’s are compatible with native platforms, Unity and Unreal. The difference is that ARKit has the edge in terms of hardware and tracking reliability, while ARCore marches ahead with its mapping and reliable recovery.

Features of ARcore/ARkit

Motion Tracking: Both development kits use the device’s motion sensors to accurately track the device position in the real-world.

Surface Detection: They have the capability to be aware and identify real time surfaces easily.

Light Estimation: The SDK can detect information about the lighting of its environment and provide the average intensity and color correction of a given camera, so that it is useful to light virtual objects under the same conditions as the environment around them, increasing the sense of realism.

Studying the Spectra

The AR spectra is growing larger day by day. But in today's scenario there is a certain inclination where AR apps are being tried and tested in a few specific areas alone. To learn and understand more about this area, a survey of the available variety of apps on the Play Store and the App Store were taken and analysed.

The survey was done both manually and automated. The consolidated data and conclusions from the survey were arrived after:

  • Testing the applications (Selected Applications based on ratings).
  • Analysing their SDK, techniques, and interaction capability.
  • Ratings, reviews, download rate.
  • Cataloguing based on their level of importance (either feature or platform).

AR App Insights Visualization

AR applications from various categories were tested and analysed on various factors to understand their distribution of creation. As mentioned, many applications had to be manually understood where the application was downloaded and analysed on their features that were useful and relevant.

Broadly, the following categories were the basis for analysing the applications for further understanding and inference:

  • Complete or Alternative platform - Whether the app uses AR as core feature or just as an alternative platform.
  • ARCore Technology - Technique used in the application like Surface Detection, Motion Tracking, Marker Based, Overlay or a combination of any.
  • Level of Interaction - The interactive capabilities with augmented objects in the application ranging from high, medium, low and no interaction.
  • Categories - Google Play and App Store categorizes application on their features provided by the app owners to promote the application. For the sake of analysing we redefined the categories based on their usability  Stickers, Games and Environment.

Some of the details available in the Google Play and App Store were considered as other factors aiding to analysis. Using an automated tool the number of downloads, ratings, number of people who rated the application and other data were obtained. On the basis of the data, the following inferences could be reached:

  • Highly rated applications are bug-free and serves the purpose.
  • Highly downloaded applications have more people's interest in terms of features and interaction.
  • Applications with less number of downloads and low rating are considered as unpopular.

The analysed data is available online, with a Sunburst Graph Visualization for better understanding, required for filtering and combining factors.  Here are four rings pointing to different dimensions of the application.

To view data:

  • Clicking on any section of ring, will fix the dimension value and expand its next level of data.
  • Hovering on the section will say the dimension name.
  • Outermost ring points the application and clicking on will take you to the Google Play site.
  • Breadcrumb is displayed on top to say the filtering hierarchy.
  • The graph legend is displayed below the visualization.

TODO: {{< ar-viz >}}


Out of 200 application explored,

  • 82.8% of the apps were using AR as core feature, whereas 17.2% of apps were using  AR as alternative platform.

In subset of apps that are using AR as core feature,

  • 64% of apps use “Surface Detection” technique for augmentation.
  • 16% of apps use Motion Tracking.
  • 8% of apps are Marker based.
  • 5% of apps augment as just camera overlay.
  • Remaining apps use combination of two or more techniques mentioned.

As our area of interest were Surface Detection and Motion Tracking, we narrowed down our analysis in these two spectra. In subset of apps that are using AR as core feature and Surface Detection,

  • 12.7% apps are highly interactive.
  • 31% are medium interactive.
  • 48% are less interactive.
  • 7% apps are not having any kind of interactions. They are just used for viewing the augmentation and lack interaction with those objects.

In the subset of complete AR apps, that have Surface Detection and high interaction,

  • 61% are games.  For example, JengaAR and AR Toys are high interactive games that use AR as core feature and implement Surface Detection technique.
  • 15% are education apps. Augmented Repair and Civilization AR are high interactive educational apps that use AR as core feature and implement Surface Detection technique.
  • 7% are tools.
  • Remaining are entertainment/stickers apps.

Overall for mobile platforms like Android\IOS,

  • AR is primarily used in Games/Entertainment apps.
  • Secondly, AR is used in utility apps like MeasureAR, Air Measure that helps in measuring the real-time distance. Though not accurate, they can give an approximate measurements of the marked area.
  • AR also has an impact on the field of Education. Apps like Math Ninja help kids to learn Math of different levels (easy/medium/hard) by interacting with AR objects.

AR is also used in other domains like