After quite some time in heads down mode I want to share with you a first glimpse at where is SSTT is heading. Obviously, the target in the demo is a sheet of paper but the internal pipeline is completely spatial. SSTT sports its own vision core with a JIT backend and a high performance math module. Everything is bundled and plug-able. Even in this early stage the tracking system is being tested across a wide range of mobile devices including smart glasses.
Some trial with a paper from Sattar et al. - with a current project I needed a faster and more robust marker system. Just a first draft with lots of room for improvement.
SSTT Bounce is a simple AR Browser, literally. It uses a WebKit based browser and adds fast AR NFT tracking to it. Rendering is done by using three.js. Beside that, Bounce can be also used as an authoring environment, it automatically detects changes in the file structure and reloads files accordingly. This allows for some nifty "in-place" editing.
Beware, its not optimized to say the least, nor otherwise safe to use. It will eat little kittens for breakfast. If you are still not scared, give it a go.
SSTT NFT is one of the tracking modules in SSTT core. Originally SSTT was conceived to replace the venereal ARToolkit as the standard tracker in osgART. Hence, the original implementation did only consider marker tracking but already abstracted tracking targets, localization and tracking on a higher level. The video below shows a prototype of SSTT with natural feature tracking using a non-optimized SURF descriptor. Due to the rather heavy weight implementation of using SURF without additional optimizations this part is not viable for the mobile version of SSTT.
SSTT mobile is a subset of SSTT core optimized for mobile devices. It currently supports only marker tracking, skin and face recognition. SSTT mobile can be used and deployed on a number of devices as listed below. SSTT mobile was originally developed for Windows Mobile 5 and 6 ported to Android, Maemo, iOS and Symbian. Only the iOS, Maemo and especially the Android port are being continued.
SSTT - (Simplified Spatial Target Tracker) is a computer vision based tracking library for Augmented and Mixed Reality applications. SSTT aims to be used for novel interaction methods in AR and therefore prioritizes robustness over speed. It is a versatile toolbox of vision based tracking modules for desktop and embedded systems such as mobile phones. It implements numerous model based approaches from traditional markers, ID based frames, rectangular textured targets, natural feature targets and object recognition with online learning.