The goal of the proposed tracking system is to track the 3D poses of a set of rigid bodies from RGB-D data streams, as obtained by Kinect, as accurate as possible:
where represents an affine 3D transformation which maps the 3D pose of an object between two frames and is the likelihood of and given and . and are the RGB and the depth information, respectively, delivered by the sensor. The tracked objects are defined as 3D point clusters, or Point Distribution Models (PDMs), , where is the total number of tracked clusters. The pose of each cluster model is given at every frame by its corresponding relative affine transformation . The position of the clusters is related to their centroids.
The tracking is initialized through tabletop object segmentation, which calculates the 3D reference model, or cluster, of the object of interest. Once is known, its shape is projected into the 2D image, where a classifier for tracking is trained on the resulted projection. Inside the tracking loop, the classifier establishes 2D correspondences between the consecutive frames and , where represents the discrete time. These 2D image matches are used to select the 3D point correspondences between the corresponding point clouds and in the Kinect data. From , an initial coarse transform for the reference cluster is calculated. Although, depending on the quality of the matches, can, to some extent, provide good tracking results, it fails to precisely map onto the current cloud . For this reason, a second transform is determined using an Iterative Closest Point (ICP) algorithm applied on non-occluded object points. Thus, the final object model transform , which tracks the pose of a 3D object between consecutive frames, can be written as:
S.M. Grigorescu, D. Pangercic and M. Beetz "2D-3D Collaborative Tracking (23CT): Towards Stable Robotic Manipulation", Proceedings of the 2012 IEEE-RSJ International Conference on Intelligent RObots and Systems IROS, Workshop on Active Semantic Perception, Vilamoura, Algarve, Portugal, October 7-12, 2012.
S.M. Grigorescu and C. Pozna "Towards a Stable Robotic Object Manipulation through 2D-3D Features Tracking", International Journal of Advanced Robotic Systems, InTech, vol. 10, no. 200, pp. 1-8, 2013.