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

Dynamic Compact Visualizations for Augmented Reality


Careless augmentation of the environment can cause information overflow, especially on small-screen devices. Simple filtering techniques reduce the amount of displayed data, but may still not lead to a comprehensible visual outcome, because the resulting layout is not considered during filtering. Compact visualization filters data by reducing redundancies and creating a layout of the remaining information. Previously, this approach was applied to create static compact explosion diagrams. We generalize the approach to work with different types of date and apply it to handle annotations, which are a major source of information in AR, and create compact layouts of annotations and annotated explosion diagrams. In addition, we demonstrate how our approach can be used in dynamic Augmented Reality environments. Therefore, we present a method to achieve interactive frame rates even on limited-resource hardware, such as mobile phones, as well as a method to automatically generate temporally coherent and scene-aware layouts.



A video containing examples and more details on our approach.


Unfiltered and cluttered visualization Dynamic Compact Visualization

(Top) Unfiltered augmentations may quickly lead to clutter and thus decrease the comprehensibility of the resulting visualization. (Bottom) Similar items have been clustered and representatives have been selected from each cluster. This allows us to reduce the amount of augmentations while still presenting an annotation to each available object class. In addition, our system allows to control the selection of representatives according to design rules. In this case, we select those annotations which most evenly distribute around the house.


Unfiltered and cluttered visualization Dynamic Compact Visualization

(Top) A naive augmentation may also result in clutter due to interfere with elements of the real world. Notice how several exploded parts, annotations and leader lines of labels interfere with the text boxes that are part of the real world poster. (Bottom) Our approach avoids these pitfalls through the combination of filter and layout techniques. Clutter is reduced by exploding and labeling only representative parts of the airplane, while the positions of text boxes have been considered in the final layout.


Unfiltered and cluttered visualization Dynamic Compact Visualization

Real world structures can be annotated automatically by applying object recognition techniques. However, the generated amount of information is unpredictable and may easily result in an overflow of augmentations. (Left) The system presents all available data including redundant labels, such as the five labels to the balustrades. (Right) After filtering the data, only a few labels remain. They have been spread over the image plane to avoid interferences with other labels during camera motion.

Dynamic Compact Visualizations for Augmented Reality. Markus Tatzgern, Denis Kalkofen, Dieter Schmalstieg, IEEE Virtual Reality Conference (VR '13), 2013


copyright (c) 2014 Graz University of Technology