Color-aware surface registration

Shape registration is fundamental to 3D object acquisition; it is used to fuse scans from multiple views. Existing algorithms mainly utilize geometric information to determine alignment, but this typically results in noticeable misalignment of textures (i.e. surface colors) when using RGB-depth cameras. We address this problem using a novel approach to color-aware registration, which takes both color and geometry into consideration simultaneously. Color information is exploited throughout the pipeline to provide more effective sampling, correspondence and alignment, in particular for surfaces with detailed textures. Our method can furthermore tackle both rigid and non-rigid registration problems (arising, for example, due to small changes in the object during scanning, or camera distortions). We demonstrate that our approach produces significantly better results than previous methods. Graphical abstractUsing rigid and non-rigid registration to correct misalignments in geometry and texture. Two input textured surfaces (a) are captured by RGB-D cameras; the camera configuration provides initial alignment (b). Successive rigid (c) and non-rigid (d) steps improve it, giving a final surface with high-quality texturesDisplay Omitted HighlightsUsing color and geometric to źnd correspondences in both rigid and non-rigid cases.Using color when determining the transformation/deformation in aligning surfaces.Proving our algorithm's effectiveness when aligning richly textured surfaces.