Face verification with balanced thresholds
Publication in refereed journal

Times Cited
Web of Science8WOS source URL (as at 31/07/2020) Click here for the latest count
Altmetrics Information

Other information
AbstractThe process of face verification is guided by a prelearned global threshold, which, however, is often inconsistent with class-specific optimal thresholds. It is, hence, beneficial to pursue a balance of the class-specific thresholds in the model-learning stage. In this paper, we present a new dimensionality reduction algorithm tailored to the verification task that ensures threshold balance. This is achieved by the following aspects. First, feasibility is guaranteed by employing an affine transformation matrix, instead of the conventional projection matrix, for dimensionality reduction, and, hence, we call the proposed algorithm threshold balanced transformation (TBT). Then, the affine transformation matrix, constrained as the product of an orthogonal matrix and a diagonal matrix, is optimized to improve the threshold balance and classification capability in an iterative manner. Unlike most algorithms for face verification which are directly transplanted from face identification literature, TBT is specifically designed for face verification and clarifies the intrinsic distinction between these two tasks. Experiments on three benchmark face databases demonstrate that TBT significantly outperforms the state-of-the-art subspace techniques for face verification.
All Author(s) ListYan SC, Xu D, Tang XO
Journal nameIEEE Transactions on Image Processing
Volume Number16
Issue Number1
Pages262 - 268
LanguagesEnglish-United Kingdom
Keywordsdimensionality reduction; face verification; subspace learning; threshold balance
Web of Science Subject CategoriesComputer Science; Computer Science, Artificial Intelligence; COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE; Engineering; Engineering, Electrical & Electronic; ENGINEERING, ELECTRICAL & ELECTRONIC

Last updated on 2020-31-07 at 23:14