SSTT - Simplified Spatial Target Tracker is a computer vision based tracking library for Augmented and Mixed Reality applications. The basic variant implements a marker based approach with a minimal set of constraints. Unlike similar libraries such as ARToolKit or the marker based StudierStube the approach of SSTT does not require thick black borders. The SSTT tracking algorithm automatically qualifies company logos or even images in books immediately as tracking targets. The following video illustrates the tracking performance.

Augmented Reality with SSTT from Hartmut Seichter on Vimeo.

Implementation

The software is highly portable and should run on a number of devices including mobile phones. It is optimized for speed and is modular in terms of the concepts used to track targets. It can utilise hybrids or sequential methods for high robustness in any lighting condition. It is implemented internally with C++ but exposes the API in pure ANSI C.

Features

SSTT has the following features

• five different tracking methods for rectangular fiducials, with hybrid and chain modes
• subpixel accurate corner detection of fiducials
• no training of fiducials necessary
• variable patch size (or auto-adaptive to the speed of the target system)
• extreme high performance on large input images (e.g. HD1080p)
• numerous filtering methods for corner detection and pose estimation
• uses spatial statistical model to optimize visibility information
• 6DOF pose estimation optimized for usage with OpenGL
• special video capture module for Windows, Windows Mobile and Mac OS X to reduce the time spent for frame retrieval

Platform

• Windows XP, Vista 64 and 32 bit
• Mac OS X 10.4.x / 10.5.x
• Linux 2.6.x with GStreamer 0.10.x
• Windows Mobile 5/6

Need more Information?

Please feel free to contact technotecture info@technotecture.com for more information.