Multi-Modal Matching Applied to Stereo

Multi-Modal Matching Applied to Stereo
复制标题

多模态匹配应用于立体声

DOI:
--
复制
发表时间:
2003
期刊:
British Machine Vision Conference
影响因子:
--
通讯作者:
N. Krüger
N. Krüger
中科院分区:
--
文献类型:
--
作者:
N. Pugeault;N. Krüger

文献摘要

被引文献

相似文献

我们引入了一种紧凑的图像信息编码,明确地将视觉信息分为几何信息(方向)、结构信息(相位和颜色)和时间信息(光流)。我们研究了这些视觉属性对大型数据集立体匹配的重要性。从这些调查中我们可以得出结论,不同属性的组合才能产生最佳结果。不同视觉属性的相对重要性的具体权重是统计确定的。
We introduce a compact coding of image information which explicitly separates visual information into geometric information (orientation) and structural information (phase and colour) and temporal information (optic flow). We investigate the importance of these visual attributes for stereo matching on a large data set. From these investigation we can conclude that it is the combination of different attributes that gives the best results. Concrete weights for the relative importance of different visual attributes are statistically determined.