Abstract. In this paper, we address the problem of surface tracking in multiple camera environments and over time sequences. In order to fully track a surface undergoing significant deformations, we cast the problem as a mesh evolution over time. Such an evolution is driven by 3D displacement fields estimated between meshes recovered independently at different time frames. Geometric and photometric information is used to identify a robust set of matching vertices. This provides a sparse displacement field that is densified over the mesh by Laplacian diffusion. In contrast to existing approaches that evolve meshes, we do not assume a known model or a fixed topology. The contribution is a novel mesh evolution based framework that allows to fully track, over long sequences, an unknown surface encountering deformations, including topological changes. Results on very challenging and publicly available image based 3D mesh sequences demonstrate the ability of our framework to efficiently recover surface motions. 1

This paper proposes a novel approach to motion capture from multiple, synchronized video streams, specifically aimed at recording dense and accurate models of the structure and motion of highly deformable surfaces such as skin, that stretches, shrinks, and shears in the midst of normal facial expressions. Solving this problem is a key step toward effective performance capture for the entertainment industry, but progress so far has been hampered by the lack of appropriate local motion and smoothness models. The main technical contribution of this paper is a novel approach to regularization adapted to nonrigid tangential deformations. Concretely, we estimate the nonrigid deformation parameters at each vertex of a surface mesh, smooth them over a local neighborhood for robustness, and use them to regularize the tangential motion estimation. To demonstrate the power of the proposed approach, we have integrated it into our previous work for markerless motion capture [9], and compared the performances of the original and new algorithms on three extremely challenging face datasets that include highly nonrigid skin deformations, wrinkles, and quickly changing expressions. Additional experiments with a dataset featuring fast-moving cloth with complex and evolving fold structures demonstrate that the adaptability of the proposed regularization scheme to nonrigid tangential motion does not hamper its robustness, since it successfully recovers the shape and motion of the cloth without overfitting it despite the absence of stretch or shear in this case. 1.

The presence of characteristic fine folds is important for modeling realistic looking virtual garments. While recent garment capture techniques are quite successful at capturing the low-frequency garment shape and motion over time, they often fail to capture the numerous high-frequency folds, reducing the realism of the reconstructed spacetime models. In our work we propose a method for reintroducing fine folds into the captured models using datadriven dynamic wrinkling. We first estimate the shape and position of folds based on the original video footage used for capture and then wrinkle the surface based on those estimates using space-time deformation. Both steps utilize the unique geometric characteristics of garments in general, and garment folds specifically, to facilitate the modeling of believable folds. We demonstrate the effectiveness of our wrinkling method on a variety of garments that have been captured using several recent techniques.

We present a novel technique for acquiring detailed facial geometry of a dynamic performance using extended spherical gradient illumination. Key to our method is a new algorithm for jointly aligning two photographs, under a gradient illumination condition and its complement, to a full-on tracking frame, providing dense temporal correspondences under changing lighting conditions. We employ a two-step algorithm to reconstruct detailed geometry for every captured frame. In the first step, we coalesce information from the gradient illumination frames to the full-on tracking frame, and form a temporally aligned photometric normal map, which is subsequently combined with dense stereo correspondences yielding a detailed geometry. In a second step, we propagate the detailed geometry back to every captured instance guided by the previously computed dense correspondences. We demonstrate reconstructed dynamic facial geometry, captured using moderate to video rates of acquisition, for every captured frame.

Abstract—In this paper we introduce a Facial Animation system using real three-dimensional models of people, acquired by a 3D scanner. We consider a dataset composed by models displaying different facial expressions and a linear interpolation technique is used to produce a smooth transition between them. For the interpolation to be applied, there must be found one-to-one correspondences between the meshes of each facial expression. Instead of focusing in the computation of dense correspondence, some points are selected and triangulation is defined, being refined by consecutive subdivisions, that compute the matchings of intermediate points. We are able to animate any model of the dataset, given its texture information for the neutral face and the geometry information for all the expressions along with the neutral face. This is made by computing matrices with the variations of every vertex when changing from the neutral face to the other expressions. The knowledge of the matrices obtained in this process makes it possible to animate other models given only the texture and geometry information of the neutral face. Furthermore, the system uses 3D reconstructed models, being capable of generating a three-dimensional facial animation from a single 2D image of a person. Also, as an extension of the system, we use artificial models that contain expressions of visemes, that are not part of the expressions of the dataset, and their displacements are applied to the real models. This allows these models to be given as input to a speech synthesis application in which the face is able to speak phrases typed by the user. Finally, we generate an average face and increase the displacements between a subject from the dataset and the average face, creating, automatically, a caricature of the subject. Keywords-Computer Graphics, Facial Animation, 3D Reconstruction. I.

We present a framework that captures and synthesizes high resolution facial geometry and performance. In order to capture highly detailed surface structures, a theory of fast normal recovery using spherical gradient illumination patterns is presented to estimate surface normal maps of an object from either its diffuse or specular reflectance, simultaneously from any viewpoints. We show that the normal map from specular reflectance yields the best record of detailed surface shape, which can be used for geometry enhancement. Moreover, the normal map from the diffuse reflectance is able to produce a good approximation of subsurface scattering. Based on the theory, two systems are developed to capture high resolution facial geometry of a static face or dynamic facial performance. The static face scanning system consists of a spherical illumination device, two singlelens reflex (SLR) cameras and a video projector. The spherical illumination device is used to cast spherical gradient patterns onto the subject. The captured spherical gradient images are then turned into surface normals of the subject. The two cameras and one projector are used to build a structured-light-assisted two-view stereo system, which acquires a moderate resolution geometry of the subject. We then use the acquired specular normal

In this paper we present a method to iteratively capture the dynamic evolution of a surface from a set of point clouds independently acquired from multi-view videos. This is done without prior knowledge on the observed shape and simply deforms the first reconstructed mesh across the sequence to fit these point clouds while preserving the local rigidity with respect to this reference pose. The deformation of this mesh is guided by control points that are randomly seeded on the surface, and around which rigid motions are locally averaged. These rigid motions are computed by iteratively re-establishing point-to-point associations between the deformed mesh and the target data in a way inspired by ICP. Our method introduces a way to account for the point dynamics when establishing these correspondences, a higher level rigidity model between the control points and a coarse-to-fine strategy that allows to fit the temporally inconsistent data more precisely. Experimental results, including quantitative analysis, on standard and challenging datasets obtained from real video sequences show the robustness and the precision of the proposed scheme. 1.

In this paper we present a new method to capture the temporal evolution of a surface from multiple videos. By contrast to most current methods, we introduce an algorithm that uses no prior of the nature of tracked surface. In addition, it does not require sparse features to constrain the deformation but only relies on strictly geometric information: a target set of 3D points and normals. Our approach is inspired by the Iterative Closest Point algorithm but handles large deformations of non-rigid surfaces. To this end, a mesh is iteratively deformed while enforcing local rigidity with respect to the reference model. This rigidity is preserved by diffusing it on local patches randomly seeded on the surface. The iterative nature of the algorithm combined with the softly enforced local rigidity allows to progressively evolve the mesh to fit the target data. The proposed method is validated and evaluated on several standard and challenging surface data sets acquired using real videos. 1.

This paper proposes a novel approach to nonrigid, markerless motion capture from synchronized video streams acquired by calibrated cameras. The instantaneous geometry of the observed scene is represented by a polyhedral mesh with fixed topology. The initial mesh is constructed in the first frame using the publicly available PMVS software for multi-view stereo [7]. Its deformation is captured by tracking its vertices over time, using two optimization processes at each frame: a local one using a rigid motion model in the neighborhood of each vertex, and a global one using a regularized nonrigid model for the whole mesh. Qualitative and quantitative experiments using seven real datasets show that our algorithm effectively handles complex nonrigid motions and severe occlusions.

Figure 1: Multiview face capture using polarized spherical gradient illumination. (a) Acquired data from five viewpoints used for stereo reconstruction. (b) Reconstructed geometry. (c) Hybrid normal rendering [Ma et al. 2007]. We present a novel process for acquiring detailed facial geometry with high resolution diffuse and specular photometric information from multiple viewpoints using polarized spherical gradient illumination. Key to our method is a new pair of linearly polarized lighting patterns which enables multiview diffuse-specular separation under a given spherical illumination condition from just two photographs. The patterns – one following lines of latitude and one following lines of longitude – allow the use of fixed linear polarizers in front of the cameras, enabling more efficient acquisition of diffuse and specular albedo and normal maps from multiple viewpoints. In a second step, we employ these albedo and normal maps as input to a novel multi-resolution adaptive domain message passing stereo reconstruction algorithm to create high resolution facial geometry. To do this, we formulate the stereo reconstruction from multiple cameras in a commonly parameterized domain for multiview reconstruction. We show competitive results consisting of high-resolution facial geometry with relightable reflectance maps using five DSLR cameras. Our technique scales well for multiview acquisition without requiring specialized camera systems for sensing multiple polarization states.