handheld augmented reality

Augmented Reality Anywhere and Anytime   


   Social Augmented Reality

   Information Presentation

   Real-Time Self Localization

   Structural Modelling

   AR Graphics

   AR Navigation

   Augmented Reality Games

   Past Projects



   Hybrid SLAM

   Panorama SLAM

   Planar SLAM

   Model-Based Tracking

   Marker-Based Tracking


Software Libraries

   Studierstube ES

   Studierstube Tracker










   Student projects


   Site map


The Handheld Augmented Reality
Project is supported by the
following institutions:



Christian Doppler Forschungsgesellschaft


Graz University of Technology




Come to ISMAR 2011



Follow us on twitter

Incremental Tracking

Tracking at least temporarily from unknown environments is very desirable since users can usually not be constrained to always point the camera straight to markers and refrain from occluding tracking targets. We developed two computationally inexpensive approaches to support marker-based global localization with incremental tracking from untrained natural features. Both techniques have been run on cell phones at interactive frame rates, and can extend the usability of markers well beyond their original purpose: If markers are temporarily lost or occluded, the incremental tracking fills in the gap until a marker is reacquired.

Incremental Tracking using Feature Following

In many applications markers are placed on a planar surface of interest that shall be augmented by the AR application. Hence there is usually texture around the marker that can be used for natural feature tracking. We exploit this fact by combining marker tracking with a feature following approach operating in a plane. As long as the marker is visible, it is treated as ground truth, and features around the marker are extracted, but not used. Since we assume that features lie in the same plane as the marker, their 3D location can directly be computed from the marker tracking. As soon as the marker tracking fails, the tracker matches the features of the current frame against those of the previous frame via template matching, and begins tracking incrementally.

Incremental Tracking using Optical Flow

Incremental tracking of orientation with inertial sensors has been shown to be highly useful for AR applications to either improve tracking robustness, or as a fallback when no other tracking approach is available. While most of today's mobile phones do not have inertial sensors, their built-in camera can be used in a similar way using optical flow. Our pixel flow detector is intended for augmenting a panoramic view of the environment. A marker is used for initially determining the current global location and viewing direction. Then the user is free to turn around observing the augmentations, while remaining in the same location.


A video showing the incremental tracker in action on a mobile found can be found in the media section (look for "Advanced Marker-Based Tracking").


copyright (c) 2014 Graz University of Technology