摘要 |
A method and an apparatus (20) for extracting feature correspondences from images are described. An image dataset, features of the images and preliminary correspondences of the features are acquired (10) as input data. At least one cluster of the features is then generated (11). In a same cluster, each feature is coupled to at least one other feature as preliminary feature correspondences. For each cluster, primary feature correspondences of the features are determined (12), including a determination of consistency measures between every two features in the cluster. |