The Information & Mutual Information Ratio for Counting Image Features and Their Matches
This work addresses feature extraction and matching for computer vision tasks like image reconstruction and recognition, but it appears incremental as it builds on classic information measures.
The paper tackled the problem of feature extraction and matching in computer vision by proposing two new image features, the Information Ratio (IR) and Mutual Information Ratio (MIR), and validated their effectiveness on INRIA Copydays and Oxford Affine Covariant Regions datasets.
Feature extraction and description is an important topic of computer vision, as it is the starting point of a number of tasks such as image reconstruction, stitching, registration, and recognition among many others. In this paper, two new image features are proposed: the Information Ratio (IR) and the Mutual Information Ratio (MIR). The IR is a feature of a single image, while the MIR describes features common across two or more images.We begin by introducing the IR and the MIR and motivate these features in an information theoretical context as the ratio of the self-information of an intensity level over the information contained over the pixels of the same intensity. Notably, the relationship of the IR and MIR with the image entropy and mutual information, classic information measures, are discussed. Finally, the effectiveness of these features is tested through feature extraction over INRIA Copydays datasets and feature matching over the Oxfords Affine Covariant Regions. These numerical evaluations validate the relevance of the IR and MIR in practical computer vision tasks