This article raises the problem of high-tech deception using video conferencing means during distance learning, which is of increased relevance due to the digitalization of the educational process, with the growth of digital literacy of young people. The article presents some methods of fraud, including a relatively new technology that is very popular: Deepfake. The article details two popular tools for replacing faces, each of which provides step-by-step instructions on how to create, configure and apply in life. The organizational methods that are presented will help teachers detect even the most disguised forgery, including future voice spoofing, and stop any attempt at deception. At the end of the article, the system is described with the help of the technologies that used to combat deepfakes, and an assessment of the danger of existing tools is given. The objectives of this article are to raise public interest in the problem of face substitution and high-tech deception in general, to create grounds for discussing the need to switch to a remote format of events, to broaden the horizons of the reader and provide him with an area for further work and research.
Distance learning has had a huge impact on the educational process of universities and schools. The main platform through which the e-leaming process at Moscow Aviation Institute (National Research University) (MAI) is carried out is LMS MAI Moodle. It should be noted that on the basis of this platform, a website with courses lms.mai.ru was created. The second in popularity and frequency of use is Microsoft Teams platform (MT). It is worth noting that thanks to the platforms for distance learning, the educational process was not only not disrupted or stopped, but was also supplemented with such advantages as autonomy and flexibility in acquiring knowledge. In connection with the coronavims pandemic, on March 17, 2020, MAI completely switched to distance learning. At the moment, the epidemiological situation in Russia is improving and universities are again conducting face-to-face classroom studies. But what was happening with two platforms mentioned above after the end of the quarantine? Students and teachers of MAI say that the use of platforms has decreased, but classes are still held in this format. We discussed the prospects for further use of MT and LMS MAI in this article.
Context and relevance. Modern trends in technology development are moving towards the transfer of computing from cloud platforms to embedded devices. In this regard, there is a growing need to determine the optimal stack of hardware and software technologies for solving computer vision problems. Goal. Determine the optimal hardware configuration for real-time computer vision tasks based on the criteria of bandwidth, latency, accuracy, and ease of integration. Methods and materials. The paper provides a direct experimental comparison of two embedded platforms for executing the YOLO11n object detection model: a single-board Orange Pi 5 computer with a built-in Rockchip RK3588S neuroprocessor (6 TOPS, INT8) and a Raspberry Pi 4B bundle with an external Google Coral Edge TPU USB accelerator (4 TOPS, INT8). The YOLO11n model was converted from a single PyTorch source file to the target formats RKNN and TensorFlow Lite (full integer quantization) using the tool chains recommended by the manufacturers. The measurements were performed under identical pretreatment conditions. Results. The Orange Pi 5 achieved a steady frame rate of 43-54 FPS (640×640 resolution) with an end-to-end delay of 19-23 ms; accuracy mAP@0.5:0.95 with INT8 quantization, it remains at ~39%. On a Raspberry Pi 4B with Edge TPU at a resolution of 448×448, the delay was 121.5 ms (~8.2 FPS), while 45% of operations are performed on the CPU, which limits scaling. Conclusions. The Orange Pi's gain in speed reaches 6-7 times at a higher resolution, the NPU demonstrates a significant load margin, and the single-board implementation eliminates the overhead of the external interface. The totality of the results allows us to recommend Orange Pi 5 as the preferred platform for building productive edge solutions with real-time object detection.
The article proposes a technology for validating autonomous navigation methods for unmanned aerial vehicles based on simultaneous mapping and localization (SLAM) algorithms. The proposed technology is based on experimental studies using real flight data and includes the design of validation scenarios, synchronization and spatial alignment of trajectories, as well as the calculation of a set of quantitative metrics characterizing global and local accuracy, drift accumulation, and the stability of the navigation solution. As a demonstration example, the technology was evaluated using the VINS-Fusion visual-inertial algorithm using recordings of real UAV flights. The obtained results show that the proposed approach allows comparing different SLAM solutions under the same conditions and identifying flight modes that are most sensitive to error accumulation.
5 October 2021 in Revista de la Universidad del Zulia
Microsoft Teams platform is a kind of messenger created on the basis of Microsoft Skype for Business in 2017. It implements the concept of a collaborative workspace and allows group members to organize joint online meetings, hold audio and video conferences and connect additional Microsoft services. Microsoft Teams implements the possibility of individual interaction between the teacher and the student through audio and video calls. In relation to the pandemic of coronavirus infection, on March 17, 2020, MAI completely switched to the distance learning mode. For the organization of distance educational process, LMS MAI website, based on the Moodle platform, is mainly used, but the range of its capabilities is limited. In order to expand opportunities and share the flow of students using the website for e-learning, many classes with students began to be held at Microsoft Teams. It has an intuitive interface and meets all the requirements of confidentiality and security: encryption, multi-factor authentication and device management are used to protect information; necessary protection mechanisms against unauthorized access are implemented. When this platform was studied a number of its advantages were identified compared to LMS Moodle platform, also there were system vulnerabilities correlated with LMS and suggestions to eliminate them.