دانلود رایگان مقاله HIRM:مدل کاهش یافته مستقل از دسته برای ویرایش مش افزایشی

Abstract

Most existing handle-based mesh deformation methods require costly re-computation for every handle set updating, namely, adding or removing of handles on the mesh surface. In this paper, we propose a reduced deformation model that is independent of handle configuration, allowing users to dynamically update the handle set without noticeable waiting time. We represent the deformation space of a mesh as propagation fields defined by only the mesh geometry, independent of the handle set. We define the propagation fields as selected eigenvectors of the Laplacian operator and adopt the transformations of isolines sampled from the fields as the deformation descriptors. In this way, the deformation descriptors are pre-computed before handle specification. During interactive manipulation, constraints generated from the handles are incorporated into the deformation system in real time. Our method therefore supports incremental mesh editing where the user can freely define different handle sets to edit different parts of the shape without waiting for long re-computation. Our reduced model is scalable since the updating time per iteration is independent of the mesh size and the number of handles. We demonstrate the effectiveness of the proposed deformation method and compare its performance with related reduced deformation models.

6. Conclusion

In this paper, we develop a handle-independent reduced model for interactive incremental mesh editing. The deformation space of the input shape is captured by the propagation fields and deformation descriptors, which are defined by only the mesh geometry and are independent of the handle configuration. The handle-independent property of our reduced model allows the updating of the deformation system to be done mostly in the background, therefore users can dynamically update the handle set in real time and edit different parts of the shape in any order, without a long waiting time for the reconstruction of the deformation system. Our proposed deformation framework achieves constant per-iterative updating time, independent of the number of handles and vertices, leading to interactive and scalable mesh editing. We believe our reduced model has great potential for other applications, such as volumetric shape deformation and skeleton-less animation.