top of page
  • Writer's pictureMAXST

Development of AR-Based Technology

Updated: Oct 26, 2023

In the early 90s, early ARs augmented AR based on a fixed location and the user's viewing direction. However, these methods had a fatal disadvantage that the environment must be set in advance, and as a result, they were not widely popularized. It was not a technology that was originally developed for popularization. In this way, AR stays in a standstill due to technical limitations.

However, over 10 years, image processing technology has developed a lot. In order to know if the image has been distorted or where it has moved, you first need to be able to distinguish certain points in the image from each other. A technology that allows you to know a specific location that can be distinguished from such an image is called feature point extraction.

Even in the same year, it is not an exaggeration to say that all people who do image processing have used OpenCV that was launched by Intel, which accelerates image processing research. (Alpha released in 2000, first 1.0.0 released in 2006) With the advancement of these underlying technologies, the foundation for AR is laid.

The Emergence of ARToolKit

With this background, ARToolKit emerged and attracted great attention. Hirokazu Kato of Japan's Nara University of Science and Technology, in his paper entitled “Marker Tracking and HMD Calibration for a Video-based Augmented Reality Conferencing System”

The method was used, but after that, it was released as open source in 2001 and used by many people.

This AR Toolkit's approach is called marker-based AR. It was difficult to do complex image tracking, so it worked by augmenting an object on top of these markers that are very easy to detect. Due to its ease, the marker is still used for camera calibration, or even in the VFX field, although the shape is different.

Likewise, it is still being used in the field of AR, but these days, the performance of devices has increased, so 2D images that are more difficult to detect can be used as a substitute for markers. Of course, augmenting an object based on an image is not called a marker tracker, but except that it is more difficult to detect than a marker, a 2D image can be used like a marker tracker because it can operate on the same principle as a marker tracker.

You can check the performance of MAXST AR SDK's Image Tracker below.

(Of course it supports marker trackers too!)

Recent Posts

See All


bottom of page