Currently static detection is the most commonly used in Android malware detection. Among them, the extraction of various features is particularly important. In analysing the data flow features of applications, researchers usually use taint analysis method to extract. However, this method lack intermediate process features. So in this paper, we analyse the features of Android components to obtain application data transfer features for complementing the application data flow features and build a more complete combination of data flow features. Based on this, we propose a new Android malicious application detection method—SUIP. This method complements the missing features based on taint analysis, and combines the LightGBM algorithm to build a detection model. Finally, we use the sample set in Virusshare for experiments. Compared with the traditional static detection method of Android malicious code, the result shows that our detection method has a high detection accuracy of 98.50%.
In this paper, a system of broadcasting football video conversion into 3D cartoon animation is designed. When a sports event is broadcasted, multiple cameras are usually deployed around the field. However, at the same time, only one camera's video is available to viewers. Viewers hope to be able to watch the game from other viewpoints. Moreover, after a major sports game, some web portals provide cartoon animations of goal events. However, this is time-consuming and tedious, and only a single viewpoint is provided. Based on the proposed object tracking methods, this paper employs computer vision and computer graphics techniques to design a system that can generate 3D cartoon animations of soccer games. This allows users to watch the game from different viewpoints.
Natural language generation (NLG) applications have gained great popularity due to the powerful deep learning techniques and large training corpus. The deployed NLG models may be stolen or used without authorization, while watermarking has become a useful tool to protect Intellectual Property (IP) of deep models. However, existing watermarking technologies using backdoors are easily detected or harmful for NLG applications. In this paper, we propose a semantic and robust watermarking scheme for NLG models that utilize unharmful phrase pairs as watermarks for IP protection. The watermarks give NLG models personal preference for some special phrase combinations. Specifically, we generate watermarks by following a semantic combination pattern and systematically augment the watermark corpus to enhance the robustness. Then, we embed these watermarks into a NLG model without misleading its original attention mechanism. We conduct extensive experiments and the results demonstrate the effectiveness, robustness, and undetectability of the proposed scheme.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.