Semi-supervised object recognition based on Connected Image Transformations
作者：
Highlights：
• A semi-supervised classifier based on transformations between images is proposed.
• Local transformations are measured by the image dissimilarity from [Keysers et al.].
• Patterns are classified using the connectivity-based distance from [Fischer et al.].
• A speedup for classifying out-of-sample patterns is provided.
• The proposed algorithm outperforms state-of-the-art semi-supervised methods.
摘要
•A semi-supervised classifier based on transformations between images is proposed.•Local transformations are measured by the image dissimilarity from [Keysers et al.].•Patterns are classified using the connectivity-based distance from [Fischer et al.].•A speedup for classifying out-of-sample patterns is provided.•The proposed algorithm outperforms state-of-the-art semi-supervised methods.
论文关键词：Semi-supervised classification,Object recognition,Connectivity,Deformation models,Low-density separation
论文评审过程：Available online 29 June 2013.
论文官网地址：https://doi.org/10.1016/j.eswa.2013.06.029
