Techniques for creating and manipulating multimedia information have progressed to the point where they can now ensure a high degree of realism. DeepFake is a generative deep learning algorithm that creates or modifies face features in a superrealistic form, in which it is difficult to distinguish between real and fake features. This technology has greatly advanced and promotes a wide range of applications in TV channels, video game industries, and cinema, such as improving visual effects in movies, as well as a variety of criminal activities, such as misinformation generation by mimicking famous people. To identify and classify DeepFakes, research in DeepFake detection using deep neural networks (DNNs) has attracted increased interest. Basically, DeepFake is the regenerated media that is obtained by injecting or replacing some information within the DNN model. In this survey, we will summarize the DeepFake detection methods in face images and videos on the basis of their results, performance, methodology used and detection type. We will review the existing types of DeepFake creation techniques and sort them into five major categories. Generally, DeepFake models are trained on DeepFake datasets and tested with experiments. Moreover, we will summarize the available DeepFake dataset trends, focusing on their improvements. Additionally, the issue of how DeepFake detection aims to generate a generalized DeepFake detection model will be analyzed. Finally, the challenges related to DeepFake creation and detection will be discussed. We hope that the knowledge encompassed in this survey will accelerate the use of deep learning in face image and video DeepFake detection methods.

Homomorphic property of public-key cryptosystems is applied for several cryptographic protocols, such as electronic cash, voting system, bidding protocols, etc. Several fingerprinting protocols also exploit the property to achieve an asymmetric system. However, their enciphering rate is extremely low and the implementation of watermarking technique is difficult. In this paper, we propose a new fingerprinting protocol applying additive homomorphic property of Okamoto-Uchiyama encryption scheme. Exploiting the property ingenuously, the enciphering rate of our fingerprinting scheme can be close to the corresponding cryptosystem. We study the problem of implementation of watermarking technique and propose a successful method to embed an encrypted information without knowing the plain value. The security can also be protected for both a buyer and a merchant in our scheme.

During the past few decades, research on digital media watermarking –initially designed for digital images with the envisioned applications of copyright protection or copy control– has significantly evolved with respect to other covers (i.e., video, audio, speech) and many more potential applications, including tamper detection, broadcast monitoring, and, more recently, fake news detection. As a result, various surveys have tried to summarize certain aspects of this research field as it has grown. This has led to more than 130 survey papers being written at different points in time, describing various parts of the scientific efforts focused on digital media watermarking. Considering the above, the aim of this paper is twofold. First, we conduct a meta-survey based on 64 selected research works, in order to summarize the most notable survey papers in this field, which allows us to “draw a map” of this research area. Second, we focus on providing the requirements for digital watermarking techniques when applied to their most recent application: detecting fake news in multimedia content. Finally, an outline of the approach taken within the DISSIMILAR (Detection of fake newS on SocIal MedIa pLAtfoRms) project for the detection of disinformation is presented.

This paper presents an improved method of the reversible data hiding for JPEG images proposed by Xuan, et al. The conventional method embeds data into the JPEG quantized 8 times 8 block DCT coefficients. In this method, we found that the blocks, located in the noisy part of the image, are not suitable for embedding data. The proposed method can judge whether a block consisted of 8 times 8 DCT coefficients is located in smooth part of image by using the DC coefficient of neighboring blocks. Our method can avoid the noisy part for embedding. It results in a better performance in terms of capacity-distortion behavior.

Purpose: With rapid increase of Android devices and application systems, there is a strong demand for Android application programmers. A lot of schools are offering Android programming courses to meet this demand. However, Android programming can be different from the conventional one because it needs interactive functions through interfaces with users, which makes the study more difficult. This paper aims to propose an Android Programming Learning Assistance System, namely, APLAS, to assist the Java-based Android programming study and education. Design/methodology/approach: By adopting the test-driven development method, APLAS is designed to achieve independent learning without the presence of teachers. Using JUnit and Robolectric, the answers from the students are automatically marked in APLAS. To cover extensive materials in Android programming, APLAS offers four stages where each stage involves several topics. Findings: To evaluate the effectiveness of APLAS, we implemented the Unit Converter assignment that covers the first two topics, namely, Basic user interface in the first stage and basic activity in the second stage. Through solving the assignment, it is expected to learn basic application development. Forty novice students of an IT department in Indonesia were asked to solve both topics separately. Originality/value: The results show that APLAS is useful and helpful for the self-study of Android programming, as they could complete codes with good execution performances.