欢迎访问《图学学报》 分享到:

图学学报

• 图形学与可视化 • 上一篇    下一篇

数据驱动的大规模水面动画合成方法

  

  • 出版日期:2014-08-30 发布日期:2015-05-05

Data-driven Synthesization for Large-scale Water Animation

  • Online:2014-08-30 Published:2015-05-05

摘要: 为了生成大规模且细节丰富的水面动画,提出了一种数据驱动的合成方法。整个
合成分为两大部分:基于视频的水面重建和三角网格拼接。首先简化了线性近似的影调恢复形
状(SFS)方法,针对满足朗伯表面假设的室外大规模水面进行重建,并根据时间和空间的连续性
进行了一系列优化处理。然后提出一种针对三角网格序列的拼接方法,根据距离最小原则对重
建的表面序列进行采样,将采样结果融合得到无缝的表面。利用以上方法,对四类水面的视频
进行重建与拼接,结果表明,合成出的水面具有较高的真实感,在规模与细节之间取得了较好
的平衡,该方法可以方便地生成满足用户需求的大规模水面动画。

关键词: 水面动画, 三维重建, 网格拼接

Abstract: A new framework is presented for synthesizing water animation with large size and rich
details. The main idea of our framework consists of two steps: reconstructing mesh model from
videos and spanning the mesh by quilting. First we simplify the shape from shading method which is
linearly approximated, and use it to reconstruct the outdoor water with large scale which satisfies the
Lambert surface assumption. The result is optimized with several smooth methods according to the
constraint of consistency in time and space. Then a quilting method is introduced for triangle mesh, in
which we sample smaller blocks from the constructing mesh based on the principle of distance
minimization and merge the blocks to get the seamless surface. The process is completed with four
kinds of water. Results demonstrate that the synthesized water animation has high sense of reality and
balance the scale with detail very well. This method can conveniently synthesize large-scale water
animation meeting use′s demand.

Key words: water animation, 3D-reconstruction, mesh quilting