![]() |
|
||
Separable Spatiotemporal Priors for Convex Reconstruction of Time-Varying 3D Point CloudsTomas Simon1, Jack Valmadre2, 3, Iain Matthews4, 1, and Yaser Sheikh1 1Carnegie Mellon University, USA
2Queensland University of Technology, Australia
3Commonwealth Scientific and Industrial Research Organisation, Australia 4Disney Research Pittsburgh, USA
Abstract. Reconstructing 3D motion data is highly under-constrained due to several common sources of data loss during measurement, such as projection, occlusion, or miscorrespondence. We present a statistical model of 3D motion data, based on the Kronecker structure of the spatiotemporal covariance of natural motion, as a prior on 3D motion. This prior is expressed as a matrix normal distribution, composed of separable and compact row and column covariances. We relate the marginals of the distribution to the shape, trajectory, and shape-trajectory models of prior art. When the marginal shape distribution is not available from training data, we show how placing a hierarchical prior over shapes results in a convex MAP solution in terms of the trace-norm. The matrix normal distribution, fit to a single sequence, outperforms state-of-the-art methods at reconstructing 3D motion data in the presence of significant data loss, while providing covariance estimates of the imputed points. Keywords: Matrix normal, trace-norm, spatiotemporal, missing data LNCS 8691, p. 204 ff. lncs@springer.com
|