Figure 1: Our algorithm learns combinations of geometric features that predict the spatial arrangement of details on surfaces. Left: Input (target) mesh without details. Center/Right: Details from each source mesh (blue) are synthesized on the target mesh (pink). AbstractThe visual richness of computer graphics applications is frequently limited by the difficulty of obtaining high-quality, detailed 3D models. This paper proposes a method for realistically transferring details (specifically, displacement maps) from existing high-quality 3D models to simple shapes that may be created with easy-to-learn modeling tools. Our key insight is to use metric learning to find a combination of geometric features that successfully predicts detail-map similarities on the source mesh; we use the learned feature combination to drive the detail transfer. The latter uses a variant of multi-resolution non-parametric texture synthesis, augmented by a high-frequency detail transfer step in texture space. We demonstrate that our technique can successfully transfer details among a variety of shapes including furniture and clothing.

We propose an approach to enhance rough 3D geometry with fine details obtained from multiple normal maps. We begin with unaligned 2D normal maps and rough geometry, and automatically optimize the alignments through 2-step iterative registration algorithm. We then map the normals onto the surface, correcting and seamlessly blending them together. Finally, we optimize the geometry to produce high-quality 3D models that incorporate the high-frequency details from the normal maps. We demonstrate that our algorithm improves upon the results produced by some well-known algorithms: Poisson surface reconstruction [1] and the algorithm proposed by Nehab et al. [2].

Diffusion Tensor Magnetic Resonance Imaging (DTI) fiber tractography is a way to reconstruct fiber tracts underlying data according to local anisotropic diffusion characteristics. Reliability of fiber tracts as a result of tractography decreases due to noise in the data, error accumulation during integration and stochastic nature of the underlying data. We proposed a new similarity measure based on point-wise correspondence between tracts. Laplacian Eigenmaps are used to embed the fiber tracts into ℜ(3) based on the new similarity measure. We compared our method with a previously proposed method, on real and phantom data, that uses a 9D feature space to measure fiber similarity and showed that the new similarity measure results in a low dimensional manifold representing the fiber bundles. We presented preliminary results demonstrating that the fibers that fall far from this manifold correspond to outliers.

No abstract