Consistent as-similar-as-possible non-isometric surface registration

• DOI: 10.1007/s00371-017-1390-9
• Publication Date: 2017-05-10T07:17:23Z
• Authors: Tao Jiang; Kun Qian; Shuang Liu; Jing Wang; Xiaosong Yang; Jianjun Zhang
• Keywords: Distortion (music); Image registration

Non-isometric surface registration, aiming to align two surfaces with different sizes and details, has been widely used in computer animation industry. Various existing registration approaches have proposed for accurate template fitting; nevertheless, challenges remain. One is how avoid the mesh distortion fold over of during transformation. The other reduce amount landmarks that be specified manually. To tackle these simultaneously, we propose a consistent as-similar-as-possible (CASAP) approach. With novel defined energy, it not only achieves discretization produce result, but also requires small number little user effort only. Besides, CASAP constrained so angles triangle meshes are preserved local scales allowed change. Extensive experimental results demonstrated effectiveness comparison state-of-the-art approaches.