Compression of 3D mesh sequences by temporal segmentation

摘要

We describe a compression method for three-dimensional animation sequences that has notable advantages over existingntechniques. We first aggregate the frame data by similarity and reorganize them into clusters, which results in the sequencensplit into several motion fragments of varying lengths. To minimize the number of clusters and obtain optimal clustering,nwe perform frame alignment, which eliminates the “global” rigid transformation from each frame data and use only “pose”nwhen evaluating the similarity between frames. We then apply principal component analysis for each cluster, from whichnwe get coordinates of corresponding frames in a reduced dimension. Because similar frames are considered, the number ofncoefficients required for each frame becomes smaller; thus, we obtain better dimension reduction for a given reconstructionnerror. Further, we perform intracluster compression based on linear coding. Because every motion fragment presents similarnframes, conventional linear predictive coding can be replaced by key frame-based linear coding to achieve minimalnreconstruction error. Results show that our method can obtain a high compression ratio, with a limited reconstruction error.nCopyright © 2013 John Wiley & Sons, Ltd.