Citation

Ong, Kyle and Ng, Kok Why and Yoong, Yih Jian (2019) 3D Shape Blending: Parts Swapping. In: 2nd International Conference on Computational Intelligence and Intelligent Systems, 23-25 Nov. 2019, AVANI Atrium Bangkok, Bangkok, Thailand.

Text 127.pdf - Published Version Restricted to Repository staff only Download (726kB)

Abstract

3D shape blending is getting more and more attention recently as it can yield numerous ideas to ease the artists/designers to strive for new concept for their product design. Besides, it is also widely extended to many other research fields such as animation that relies on the automated process of interpolating a 3D model based on user-defined weights or anchors. Such a task will take a lot of effort and time if it is to be done manually. In this paper, we propose to auto-segment the object into several meaningful features. The object will first be converted into 1D skeletal for obtaining the main features. Slinky-based segmentation method is applied to separate each semantic feature. The model then is partitioned in parts and is blended (swapped) from the source to the target model. The final result shows that the blended model successfully modify the parts of a model but reveal some gaps which can be patched easily by connecting the neighboring vertices of both the source and target model.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Shape Blending, Metamorphosis, 3D Shape Creation, Topology Variation, 3D Shape Segmentation.
Subjects:	Q Science > QA Mathematics > QA440-699 Geometry. Trigonometry. Topology
Divisions:	Faculty of Computing and Informatics (FCI)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	15 Oct 2021 06:52
Last Modified:	15 Oct 2021 06:52
URII:	http://shdl.mmu.edu.my/id/eprint/9556

Downloads

Edit (login required)