Facial expression verification has been extensively exploited due to its wide application in affective computing, robotic vision, man-machine interaction and medical diagnosis. With the recent development of Internet-of-Things (IoT), there is a need of mobile-targeted facial expression verification, where face scrambling has been proposed for privacy protection during image/video distribution over public network. Consequently, facial expression verification needs to be carried out in a scrambled domain, bringing out new challenges in facial expression recognition. An immediate impact from face scrambling is that conventional semantic facial components become not identifiable, and 3D face models cannot be clearly fitted to a scrambled image. Hence, the classical facial action coding system cannot be applied to facial expression recognition in the scrambled domain. To cope with chaotic signals from face scrambling, this paper proposes an new approach – Many Graph Embedding (MGE) to discover discriminative patterns from the subspaces of chaotic patterns, where the facial expression recognition is carried out as a fuzzy combination from many graph embedding. In our experiments, the proposed MGE was evaluated on three scrambled facial expression datasets: JAFFE, MUG and CK++. The benchmark results demonstrated that the proposed method is able to improve the recognition accuracy, making our method a promising candidate for the scrambled facial expression recognition in the emerging privacy-protected IoT applications

Image steganography is used to hide a secret image inside a cover image in plain sight. Traditionally, the secret data is converted into binary bits and the cover image is manipulated statistically to embed the secret binary bits. Overloading the cover image may lead to distortions and the secret information may become visible. Hence the hiding capacity of the traditional methods are limited. In this paper, a lightweight yet simple deep convolutional autoencoder architecture is proposed to embed a secret image inside a cover image as well as to extract the embedded secret image from the stego image. The proposed method is evaluated using three datasets -COCO, CelebA and ImageNet. Peak Signal-to-Noise Ratio, hiding capacity and imperceptibility results on the test set are used to measure the performance. The proposed method has been evaluated using various images including Lena, airplane, baboon and peppers and compared against other traditional image steganography methods. The experimental results have demonstrated that the proposed method has higher hiding capacity, security and robustness, and imperceptibility performances than other deep learning image steganography methods.

Abstract-Real face recognition is a challenging problem especially when face images are subject to distortions. This paper presents an approach to tackle partial occlusion distortions present in real face recognition using a single training sample per person. First, original images are partitioned into multiple blocks and Local Binary Patterns are applied as a local descriptor on each block separately. Then, a dimensionality reduction of the resulting descriptors is carried out using Kernel Principle Component Analysis. Once done, a random sampling method is used to select patches at random and hence build several sub-SVM classifiers. Finally, the results from each sub-classifier are combined in order to increase the recognition performance. To demonstrate the usefulness of the approach, experiments were carried on the AR Face Database and obtained results have shown the effectiveness of our technique.

Steganography is the art of transmitting hidden messages through a cover object without raising any suspicion. In contrast, steganalysis is the science of detecting the presence of hidden information and a significant amount of research has been focused on multimedia steganalysis. In this paper, a video steganalysis method is proposed to detect the presence of hidden data by analysing the structure of the motion vectors in the compressed video data. The proposed method is based on the classification of features extracted from the morphology of the motion vector map. The proposed method has been evaluated on a large dataset of short videos with variable resolution and quality and the results suggest the effectiveness of the proposed modelling scheme.

No abstract