Example-based mesh deformation techniques produce natural and realistic shapes by learning the space of deformations from\r\nexamples. However, skeleton-based methods cannot manipulate a global mesh structure naturally, whereas the mesh-based\r\napproaches based on a translational control do not allow the user to edit a local mesh structure intuitively. This paper presents\r\nan example-driven mesh editing framework that achieves both global and local pose manipulations. The proposed system is built\r\nwith a surface deformation method based on a two-step linear optimization technique and achieves direct manipulations of a\r\nmodel surface using translational and rotational controls. With the translational control, the user can create a model in natural\r\nposes easily. The rotational control can adjust the local pose intuitively by bending and twisting.We encode example deformations\r\nwith a rotation-invariant mesh representation which handles large rotations in examples. To incorporate example deformations,\r\nwe infer a pose from the handle translations/rotations and perform pose space interpolation, thereby avoiding involved nonlinear\r\noptimization. With the two-step linear approach combined with the proposed multiresolution deformation method, we can edit\r\nmodels at interactive rates without losing important deformation effects such as muscle bulging.
Loading....