摘要
本文提出了一种新颖的基于改进BP算法的中间视合成方法。方法中首先改进了BP(Belief Propagation)算法来提高视差估计的正确率,然后在视图插值中设计了新的插值策略,在区域分类的基础上有针对地进行中间视合成。实验结果表明,采用本文提出的中间视合成方法可以获得具有很好视觉效果的虚拟视图。
A novel intermediate view synthesis method based on improved BP （ Belief Propagation） algorithm is presented in this paper. First, BP algorithm is improved to increase the correct rate of the disparity estimation. Then, in view interpolation, new interpolation strategy is designed and different strategy is applied according to the characteristic of each kind of region. Experimental results show that it can synthesize virtual images with high quality.
出处
《信号处理》
CSCD
北大核心
2010年第3期347-353,共7页
Journal of Signal Processing
基金
国家自然科学基金(No:60872073)
江苏省自然科学基金(BK2008291)
教育部博士点基金(No:20050286001)资助课题
关键词
视差估计
BP算法
视图插值
中间视合成
disparity estimation
BP algorithm
view interpolation
intermediate view synthesis
