Data Uncertainty in Face Recognition

Abstract

The image of a face varies with the illumination, pose, and facial expression, thus we say that a single face image is of high uncertainty for representing the face. In this sense, a face image is just an observation and it should not be considered as the absolutely accurate representation of the face. As more face images from the same person provide more observations of the face, more face images may be useful for reducing the uncertainty of the representation of the face and improving the accuracy of face recognition. However, in a real world face recognition system, a subject usually has only a limited number of available face images and thus there is high uncertainty. In this paper, we attempt to improve the face recognition accuracy by reducing the uncertainty. First, we reduce the uncertainty of the face representation by synthesizing the virtual training samples. Then, we select useful training samples that are similar to the test sample from the set of all the original and synthesized virtual training samples. Moreover, we state a theorem that determines the upper bound of the number of useful training samples. Finally, we devise a representation approach based on the selected useful training samples to perform face recognition. Experimental results on five widely used face databases demonstrate that our proposed approach can not only obtain a high face recognition accuracy, but also has a lower computational complexity than the other state-of-the-art approaches.

Dynamic dictionary optimization for sparse-representation-based face classification using local difference images

“…These images were captured over two sessions. The selected subset has also been widely used in previous studies [46,49,50]. Each image was down-sampled to …”

Heterogeneous Face Recognition by Margin-Based Cross-Modality Metric Learning

“…The synthesis-based methods map the data of one modality into another by synthesizing [8]. Related work includes synthesizing sketches from photograph and then comparing synthesized images with sketches drawn by artists [3], [4], [9]- [11].…”

The fusion of original and symmetric virtual images for image preprocessing in face recognition and collaborative representation based classification

“…[22][23][24] To this end, various improved methods have been proposed, in which the most simple but effective methods are the virtual sample based methods. [25][26][27][28][29][30] Generally, for face recognition, the more training samples are used, the higher recognition accuracy will be. This is mainly because the face of a person can be better described by more samples.…”

“…[20,31,32] For example, considering that single face image is a wide range of uncertainty for representing the face, Xu et al proposed to reduce this uncertainty by viewing the mean of two face images as virtual training samples. [27] Ryu et al [28] exploited the distribution of the virtual training samples which are generated from the given training set. Liu et al also represented every single image by some synthesized (shifted) samples.…”