摘要： 提出一种基于三角形网格的帧频提升方法。对当前帧做冗余小波变换提取特征节点，基于特征节点利用Delaunay三角剖分将当前帧划分为三角形网格，在网格确定的搜索范围内估计特征节点的运动矢量，利用特征节点的运动矢量和参考帧生成内插帧。实验结果表明，该方法比传统的运动帧频提升方法有更好的效果。

关键词: 特征节点, Delaunay三角剖分, 运动估计, 帧频提升

Abstract: A new Frame Rate Up-Conversion(FRUC) approach based on triangulation mesh is proposed. The algorithm constructs a triangular mesh based on the identified nodes of the current frame, which is extracted by implementing the redundant wavelet transform. It gets the motion vectors of the identified nodes by estimating displacement of such nodes from the previous frame with respect to the current frame in the triangular mesh. The new frame is rendered based on the reference frames and the estimated displacement of nodes. Experimental results show that the proposed algorithm provides better performance than conventional motion-compensated interpolation approach.

Key words: identified node, Delaunay triangulation, motion estimation, Frame Rate Up-Conversion(FRUC)

中图分类号: 

• TN911.73