Difference between revisions of "Spatio-temporal Segmentation for the Similarity Measurement of Deforming Meshes"
(/* Luo G., Deng, Z., Jin, X. Zhao, X., Zeng, W., Xie, W. Seo, H., “Spatio-temporal Segmentation based Adaptive Compression of Dynamic Mesh Sequences”, ACM Trans. Multimedia Computing, Communications, and Applications, accepted for publication, 2019...) |
|||
Line 13: | Line 13: | ||
== Papers == | == Papers == | ||
− | ===== Luo G., Deng, Z., Jin, X. Zhao, X., Zeng, W., Xie, W. Seo, H., “Spatio-temporal Segmentation based Adaptive Compression of Dynamic Mesh Sequences”, ACM Trans. Multimedia Computing, Communications, and Applications, | + | ===== Luo G., Deng, Z., Jin, X. Zhao, X., Zeng, W., Xie, W. Seo, H., “Spatio-temporal Segmentation based Adaptive Compression of Dynamic Mesh Sequences”, ACM Trans. Multimedia Computing, Communications, and Applications, Vol.16, No.1, Article No. 14, 2020. ===== |
===== Luo G., Zeng W., Zhao X., Deng Z., Jin X. and Seo H., “3D mesh animation compression based on adaptive spatio-temporal segmentation”, Article No. 10, ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games, 2019 (May), Montreal, Canada. ===== | ===== Luo G., Zeng W., Zhao X., Deng Z., Jin X. and Seo H., “3D mesh animation compression based on adaptive spatio-temporal segmentation”, Article No. 10, ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games, 2019 (May), Montreal, Canada. ===== |
Latest revision as of 19:23, 21 March 2020
Contributors
- Guoliang Luo, Frederic Cordier, Hyewon Seo (University of Strasbourg)
- Wei Zeng, Wenqiang Xie (Jiangxi Normal University, China)
- Guoliang Luo, Xin Zhao (East China Jiaotong University, China)
- Zhigang Deng (Univ. Houston, USA), Xiaogang Jin (Zhejiang University, China)
Abstract
Although there have been a large body of works on computing the similarity of static shapes, similarity judgments on deforming meshes are not studied well. In this study, we investigate a similarity measurement method for comparing two deforming meshes. Our algorithm uses the degree of deformation to binarily label each triangle in deforming mesh in the spatio-temporal domain, which is then encoded in a form of evolving graphs (EG) to obtain a compact representation of the given motion. Based on EG representation, we further developed a motion similarity measure between two deforming meshes, which we formulate as a graph matching problem. In this formulation, we have identified, and proposed solution to, two challenging problems: First, comparison of graphs that change their topologies over time is a difficult problem, for which few attempts have been made. Second, the evolving graph representation tend to be noisy.
More recently, we have searched for an exploitation of the spatio-temporal coherency within a segment towards a compact represent an animation mesh. Given a mesh on which a motion-driven spatio-temporal segmentation has been computed, we perform PCA-based compression on each spatio-temporal segment. Since our algorithm is designed to exploit both temporal and spatial segmentation redundancies by adaptively determining segmentation boundaries, it shows a significantly better performance over other comparable compression methods, as expected.