As a classic topic, the performance of face recognition has been greatly improved by deep neural network algorithms when a dataset is large. However, when face data are insufficient as in practical Internet of Things (IoT) applications and captured by IoT devices under the same intra-subject variation, both data quantity and quality bring big challenges to construct a model or representation, and most of the time it becomes infeasible to build a deep neural network model. This paper proposes a Sparse Individual Low-rank component-based Representation (SILR) such that the representation of testing images can be based on individual subjects’ low-rank component. Theoretically, we put the l2-norm constraint on intra-subject coefficients to represent testing images, thus making intra-subject coefficients dense. Hence, we alleviate the impact of an undersampled training dataset and its same inter-subject variation on classification performance. We solve a convex minimization problem in polynomial time via an Augmented Lagrange Multiplier scheme to get the solution of the proposed SILR. The scheme can reduce the influences from the same inter-subject variation and contribute to an accurate recognition of the undersampled training dataset. We adopt sparse individual low-rank component representation and minimum reconstruction residual to recognize testing images. Extensive results on benchmark face databases and many non-standard datasets show that the proposed SILR is better than the other state-of-the-art methods for face recognition.
All Science Journal Classification (ASJC) codes
- Signal Processing
- Information Systems
- Hardware and Architecture
- Computer Science Applications
- Computer Networks and Communications
- Face recognition
- Face recognition
- Feature extraction
- Internet of Things
- IoT-based systems.
- Training data
- sparse individual low-rank component representation
- sparse representation