Heterogeneous Face Recognition by Margin-Based Cross-Modality Metric Learning
- 1 June 2018
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cybernetics
- Vol. 48 (6), 1814-1826
- https://doi.org/10.1109/TCYB.2017.2715660
Abstract
Heterogeneous face recognition deals with matching face images from different modalities or sources. The main challenge lies in cross-modal differences and variations and the goal is to make cross-modality separation among subjects. A margin-based cross-modality metric learning ((MCML)-L-2) method is proposed to address the problem. A cross-modality metric is defined in a common subspace where samples of two different modalities are mapped and measured. The objective is to learn such metrics that satisfy the following two constraints. The first minimizes pairwise, intrapersonal cross-modality distances. The second forces a margin between subject specific intrapersonal and interpersonal cross-modality distances. This is achieved by defining a hinge loss on triplet-based distance constraints for efficient optimization. It allows the proposed method to focus more on optimizing distances of those subjects whose intrapersonal and interpersonal distances are hard to separate. The proposed method is further extended to a kernelized (MCML)-L-2 ((KMCML)-L-2). Both methods have been evaluated on an ID card face dataset and two other cross-modality benchmark datasets. Various feature extraction methods have also been incorporated in the study, including recent deep learned features. In extensive experiments and comparisons with the state-of-the-art methods, the (MCML)-L-2 and (KMCML)-L-2 methods achieved marked improvements in most cases.Keywords
Funding Information
- National Science Foundation of China (61432008, 61673203, 61603193)
- Young Elite Scientists Sponsorship Program by CAST (YESS 20160035)
- Collaborative Innovation Center of Novel Software Technology and Industrialization
- China Scholarship Council through the University of Manchester
This publication has 37 references indexed in Scilit:
- Local discriminative distance metrics ensemble learningPattern Recognition, 2013
- Multiview Metric Learning with Global Consistency and Local SmoothnessACM Transactions on Intelligent Systems and Technology, 2012
- Speed up kernel discriminant analysisThe VLDB Journal, 2010
- Distinctive Image Features from Scale-Invariant KeypointsInternational Journal of Computer Vision, 2004
- Face Sketch RecognitionIEEE Transactions on Circuits and Systems for Video Technology, 2004
- Gabor feature based classification using the enhanced fisher linear discriminant model for face recognitionIEEE Transactions on Image Processing, 2002
- KERNEL AND NONLINEAR CANONICAL CORRELATION ANALYSISInternational Journal of Neural Systems, 2000
- The FERET evaluation methodology for face-recognition algorithmsIeee Transactions On Pattern Analysis and Machine Intelligence, 2000
- Eigenfaces vs. Fisherfaces: recognition using class specific linear projectionIeee Transactions On Pattern Analysis and Machine Intelligence, 1997
- RELATIONS BETWEEN TWO SETS OF VARIATESBiometrika, 1936