Skip to Main Content
The goal of 3D object compression is to reduce the storage cost of 3D models. A 3D animation model comprises many 3D static models. Hence, in order to display a realistic animation model, many triangles are required. However, the higher storage-cost models have higher computational costs. Progressive compression is exploited to reduce the storage cost and computational cost of 3D models. However, progressive compression easily leads to the distortion of the 3D models. This study proposes a feature-sensitive progressive compression method to compress the 3D models. This method uses the temporal discrete shape operator (TDSO) to analyze the animated surface variations. It can reduce the number of triangles in the 3D animation models, and simultaneously it does not increase the distortion caused by compression. Experimentally, this study compares root mean square (RMS) errors in this method with those of the principal component analysis (PCA) compression method. Experimental results show that our approach can preserve more contour features than the PCA approach at the same compression ratio.