Cage-Based Tracking for Performance Animation

Abstract

Full body performance capture is a promising emerging technology that has been intensively studied in Computer Graphics and Computer Vision over the last decade. Highly-detailed performance animations are easier to obtain using existing multiple views platforms, markerless capture and 3D laser scanner. In this paper, we investigate the feasibility of extracting optimal reduced animation parameters without requiring an underlying rigid kinematic structure. This paper explores the potential of introducing harmonic cage-based linear estimation and deformation as post-process of current performance capture techniques used in 3D time-varying scene capture technology. We propose the first algorithm for performing cage-based tracking across time for vision and virtual reality applications. The main advantages of our novel approach are its linear single pass estimation of the desired surface, easy-to-reuse output cage sequences and reduction in storage size of animations. Our results show that estimated parameters allow a sufficient silhouette-consistent generation of the enclosed mesh under sparse frame-to-frame animation constraints and large deformation.

A lidar-based method for the detection and counting of Powered Two Wheelers

“…Li et al [11] propose a tensor-based method for a general purpose feature detector but this method constructs a scene of the static environment which is not our case study. Savoye et al [12] suggest an algorithm for performing cage-based tracking over time to visualize 3D data and for virtual reality applications but this method needs multiple sensors which is not our case as we work with a single-plane lidar. Lafarge et al [13] propose to construct the static urban scenes using lidar but do not construct dynamic objects.…”

EC-CageR: Error controllable cage reverse for animated meshes

“…Thiery et al [2] avoided the degeneracy by extracting a sub-matrix of Φ with maximal absolute determinant, while discarding other constraints. Savoye and Franco [25] prevented the degeneracy by preserving the details of the reference cage using Laplacian coordinates.…”

“…Savoye et al [1,25] presented a stable formulation to fit a cage sequence to animated meshes by preserving Laplacian coordinates of the initial cage. Noticing that this method may lead to shrinkage and distortion because the differential coordinates are not rotation invariant, Thiery et al [2] employed a maximum volume submatrices technique to select a subset of vertices on the pose to be fitted as position constraints and therefore converted an overdetermined problem into a due restraint problem.…”

Cage-Based Motion Recovery Using Manifold Learning

“…Tracking mesh data using a cage for barycentriccoordinates has first been proposed by [16] and recently improved by [18]. Unlike these works, we make no assumption on the temporal coherence of the input 3D data.…”

“…Another major difference is in the regularization strategy. In [16], the cage deformation was constrained by a Laplacian smoothing. It has been reported that it did not prevent the cage from collapsing when tracking long sequences.…”