Fuzzy match index for scale-invariant feature transform (SIFT) features with application to face recognition with weak supervision

• Published in: IET image processing Vol. 9; no. 11; pp. 951 - 958
• Main Authors: Susan, Seba, Jain, Abhishek, Sharma, Aakash, Verma, Shikhar, Jain, Siddhant
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 01.11.2015
• Subjects: Caltech‐256 face database; Decision making; decision‐making process; Entropy; Face recognition; facial features; feature entropy weights; FERET database; Fuzzy; Fuzzy logic; fuzzy match index; Fuzzy set theory; fuzzy SIFT classifier; fuzzy SIFT descriptor matching tool; image matching; labelled faces; robust face recognition; scale‐invariant feature transform features; SIFT features; soft decision‐making; Special Issue: Machine Learning in Image Processing; test SIFT keypoints; Transforms; uncropped images; weak supervision; wild dataset; robust face recognition; wild dataset; scale-invariant feature transform features; test SIFT keypoints; fuzzy set theory; fuzzy match index; transforms; fuzzy SIFT descriptor matching tool; FERET database; image matching; facial features; uncropped images; weak supervision; fuzzy SIFT classifier; entropy; labelled faces; decision making; decision-making process; soft decision-making; face recognition; SIFT features; Caltech-256 face database; feature entropy weights
• ISSN: 1751-9659; 1751-9667
• DOI: 10.1049/iet-ipr.2014.0670

A fuzzy match index for scale-invariant feature transform (SIFT) features is proposed in this study that cumulatively involves all the test SIFT keypoints in the decision-making process. The new fuzzy SIFT classifier is adapted successfully for robust face recognition from complex backgrounds without using any face cropping tools and using only a single training template. The further incorporation of entropy weights ensures that the facial features have a greater role in the soft decision-making as compared with the background features. The highlights of the authors’ work are: (i) The development of a novel highly efficient fuzzy SIFT descriptor matching tool; (ii) incorporation of feature entropy weights to highlight the contribution of facial features; (iii) application to robust face recognition from uncropped images having diverse backgrounds with a single template for each subject. The authors thus allow for weak supervision of the face recognition experiment and obtain high accuracy for 20 subjects of the CALTECH-256 face database, 133 subjects of the labelled faces for the wild dataset and 994 subjects of the FERET database, with state-of-the-art comparisons indicating the supremacy of the authors’ approach.