Release of Bayesian Articulated Object Tracking Libraries
Robust and real-time Bayesian articulated object tracking methods, implemented in C++ and CUDA.
We release open-source code and data sets on Bayesian articulated object tracking. The library contains approaches towards problems ranging from single object tracking to full robot arm pose estimation. The data sets allow the quantitative evaluation of alternative approaches thanks to accurate ground-truth annotations.
For more information on the research behind this library, please refer to the project page on Probabilistic Object and Manipulator Tracking. Part of the work that went into this library is nominated for the Best Vision paper at ICRA ‘17.
Links:
The following video visualizes the performance of our arm tracking method.
The following video visualizes the performance of our method for probabilistic object tracking.
