Human Motion Capture Driven by Orientation Measurements

Abstract

Motion-capture techniques are rarely based on orientation measurements for two main reasons: (1) optical motion-capture systems are designed for tracking object position rather than their orientation (which can be deduced from several trackers), (2) known animation techniques, like inverse kinematics or geometric algorithms, require position targets constantly, but orientation inputs only occasionally. We propose a complete human motion-capture technique based essentially on orientation measurements. The position measurement is used only for recovering the global position of the performer. This method allows fast tracking of human gestures for interactive applications as well as high rate recording. Several motion-capture optimizations, including the multijoint technique, improve the posture realism. This work is well suited for magnetic-based systems that rely more on orientation registration (in our environment) than position measurements that necessitate difficult system calibration.

A pervasive gesture-driven augmented reality prototype using wireless sensor body area networks

“…However, traditional HMM-based gesture recognition systems require a large number of parameters to be trained in order to give satisfying recognition results. In particular, an n-state HMM requires n 2 parameters to be trained for the transition probability matrix, which limits its usability in environments where training data is limited [4] [9]. The reduced model that was used in our system uses a constant number of parameters for each state to determine transition probabilities between all states.…”

Full-body performance animation with Sequential Inverse Kinematics

“…There have been previous studies in which known orientations are used for motion reconstruction such as those of Monheit and Badler [29] and Molet et al [30] for the spine, Zordan and Hodgins [31] for the upper-body, and Badler et al [32] and Semwal et al [33] for the full-body. Nevertheless we are more interested in adjusting the poses only to known positions instead of orientations, since it is more intuitive for an animator to situate them or for a motion capture system to track them.…”

Computer puppetry

“…There will be some level of mismatching even for characters that have the same size and proportion as the performer, since we simplify the real human by a hierarchy of rigid bodies. One approach to performance animation, described by Molet et al [1996Molet et al [ , 1999, models the character to be as similar to the performer as possible. Bodenheimer et al [1997] presented a way to determine the segment lengths of a character that best fit the captured motion data while discarding outliers in these data by a robust estimation technique.…”