Optimizing octree motion representation for 3D animation
Proceedings of the 44th annual Southeast regional conference
Adaptive geometry compression based on four-point interpolatory subdivision schemes with labels
International Journal of Computer Mathematics - Computer Vision and Pattern Recognition
Single-rate near lossless compression of animated geometry
Computer-Aided Design
Technologies for 3D mesh compression: A survey
Journal of Visual Communication and Image Representation
Shape-based simplification for 3d animation models using shape operator sequences
CIT'10 Proceedings of the 4th international conference on Communications and information technology
Improved prediction methods for scalable predictive animated mesh compression
Journal of Visual Communication and Image Representation
Optimised mesh traversal for dynamic mesh compression
Graphical Models
Adaptive geometry compression based on 4-point interpolatory subdivision schemes
IWICPAS'06 Proceedings of the 2006 Advances in Machine Vision, Image Processing, and Pattern Analysis international conference on Intelligent Computing in Pattern Analysis/Synthesis
The alpha parallelogram predictor: A lossless compression method for motion capture data
Information Sciences: an International Journal
Hi-index | 0.00 |
Geometry compression is the compression of the 3D geometric data that provides acomputer graphics system with the scene description necessary to render images.Geometric data is quite large and, therefore, needs effective compression methods todecrease the transmission and storage bit requirements. A large amount of research hasfocused on static geometry compression, but only limited research has addressedanimated geometry compression, the compression of temporal sequences of geometrydata. In this paper, we propose an octree-based motion representation method that can beapplied to compress animated geometric data. In our approach, 3D animated sequencescan be represented with a compression factor of over 100, with slight losses in animationquality. We focus on compressing vertex positions for all the frames. In our approach weonly need to use two consecutive frames to generate a small set of motion vectors thatrepresent the motion from the previous frame to the current frame. The motion vectorsare used to predict the vertex positions for each frame except the first frame. The processgenerates a hierarchical octree motion representation for each frame. Quantization and anadaptive arithmetic coder are used to achieve further data reduction. The simple andefficient decompression of this approach makes it suitable for real time applications.