A survey of feature matching methods

• Published in: IET image processing 2024-05, Vol.18 (6), p.1385-1410
• Main Authors: Huang, Qian, Guo, Xiaotong, Wang, Yiming, Sun, Huashan, Yang, Lijie
• Format: Article
• Language: English
• Subjects: feature extraction; image processing; learning (artificial intelligence)
• ISSN: 1751-9659; 1751-9667
• DOI: 10.1049/ipr2.13032

Feature matching plays a crucial role in computer vision, with applications in visual localization, simultaneous localization and mapping (SLAM), image stitching, and more. It establishes correspondences between sets of feature points from multiple images, enabling various tasks. Over the years, feature matching has witnessed significant development, with an increasing number of methods being applied. However, different methods exhibit different degrees of applicability in different scenarios and requirements due to their different rationales. To cope with these issues, a comprehensive analysis and comparison of matching methods are essential. Existing reviews often lack coverage of deep learning models and focus more on feature detection and description, neglecting the matching process. This survey investigates feature detection, description, and matching techniques within the feature‐based image‐matching pipeline. Representative methods, their mechanisms, and application scenarios are also briefly introduced. In addition, comprehensive evaluations of classical and state‐of‐the‐art methods are conducted through extensive experiments on representative datasets. Particularly, matching‐based applications are compared to fully demonstrate the advantages of the methods. Lastly, this survey highlights current problems and development directions in matching methods, serving as a reference for researchers in the field. Following the feature‐based image matching pipeline, we provide a deep investigation into feature detection, description, and matching techniques. And we briefly introduce several representative methods with their mechanisms, scenarios of application, etc. Then we provide a comprehensive evaluation of these classical and latest methods by conducting extensive experiments on representative datasets.