Automatic Hand-Over Animation using Principle Component Analysis

Abstract

This paper introduces a method for producing high quality hand motion using a small number of markers. The proposed "handover" animation technique constructs joint angle trajectories with the help of a reference database. Utilizing principle component analysis (PCA) applied to the database, the system automatically determines the sparse marker set to record. Further, to produce hand animation, PCA is used along with a locally weighted regression (LWR) model to reconstruct joint angles. The resulting animation is a full-resolution hand which reflects the original motion without the need for capturing a full marker set. Comparing the technique to other methods reveals improvement over the state of the art in terms of the marker set selection. In addition, the results highlight the ability to generalize the motion synthesized, both by extending the use of a single reference database to new motions, and from distinct reference datasets, over a variety of freehand motions.

Design and evaluation of reduced marker layouts for hand motion capture

“…A comparison of our marker layout optimization method with the marker subset selection approach of Wheatland et al is shown in Figure . A crucial aspect to note regarding this comparison is that the two methods are based on different marker layout generation paradigms.…”

“…Figure shows that the results of the subset selection method are strongly influenced by the choice of the base marker layout. As the method of Wheatland et al is based on computing an importance ranking for the base markers based on their positional trajectories, the selected marker layouts are clustered around the areas of the hand that move the most in the considered hand motion. In contrast, our method is sensitive to the hand kinematics and the subspace model employed in our approach, which produces layouts that are well suited for subspace‐constrained IK, as opposed to the data‐driven regression approach of Wheatland et al…”

“…In contrast, Chang et al determine the most important markers in a reduced marker set for grasp motion recognition by using supervised feature selection based on the prediction accuracy of grasp classifiers. In other works, a data‐driven approach for hand motion reconstruction from sparse marker sets was used, where motions are synthesized by finding database postures that most resemble the low‐dimensional input. Wheatland et al computed a subset of an initial full marker set by performing principal component analysis (PCA) on the marker trajectories and selecting the most influential ones.…”

“…In other works, a data‐driven approach for hand motion reconstruction from sparse marker sets was used, where motions are synthesized by finding database postures that most resemble the low‐dimensional input. Wheatland et al computed a subset of an initial full marker set by performing principal component analysis (PCA) on the marker trajectories and selecting the most influential ones. Our method differs from theirs in two significant aspects: First, our IK‐based approach allows for the recovery of hand articulations that are not present in the prior database, and second, we determine reduced marker layouts in a bottom‐up way based on the PCA of joint angles, which explicitly captures the correlations and redundancies present within hand kinematics, unlike positional marker trajectories.…”

Assessing the Impact of Hand Motion on Virtual Character Personality

“…Hand animation for virtual characters can be reconstructed from image based [Zhao et al 2012;, marker based [Kang et al 2012;Wheatland et al 2013], and glove based [Wang and Neff 2013;Huenerfauth and Lu 2010;Griffin et al 2000;Hu et al 2004;Menon et al 2003;Steffen et al 2011] motion capture techniques, or synthesized using physics knowledge [Zhao et al 2013;Liu 2008;2009;Ye and Liu 2012], data-driven [Jörg et al 2012;ElKoura and Singh 2003] or rule-based methods [Zhu et al 2012]. However, despite all the hand motion acquisition and generation research, little hand motion perception work has been conducted.…”