With the advent of mobile robots in commercial applications, the problem of path-planning has acquired significant attention from the research community. An optimal path for a mobile robot is measured by various factors such as path length, collision-free space, execution time, and the total number of turns. MEA* is an efficient variation of A* for optimal path-planning of mobile robots. RRT*-AB is a sampling-based planner with rapid convergence rate, and improved time and space requirements than other sampling-based methods such as RRT*. The purpose of this paper is the review and performance comparison of these planners based on metrics, i.e., path length, execution time, and memory requirements. All planners are tested in structured and complex unstructured environments cluttered with obstacles. Performance plots and statistical analysis have shown that MEA* requires less memory and computational time than other planners. These advantages of MEA* make it suitable for off-line applications using small robots with constrained power and memory resources. Moreover, performance plots of path length of MEA* is comparable to RRT*-AB with less execution time in the 2D environment. However, RRT*-AB will outperform MEA* in high-dimensional problems because of its inherited suitability for complex problems.

Mobile robots have various applications in agriculture, autonomous cars, industrial automation, planetary exploration, security, and surveillance. The generation of the optimal smooth path is a significant aspect of mobile robotics. An optimal path for a mobile robot is measured by various factors such as path length, path smoothness, collision-free curve, execution time, and the total number of turns. However, most of the planners generate a non-smooth less optimal and linear piecewise path. Post processing smoothing is applied at the cost of increase in path length. Moreover, current research on post-processing path smoothing techniques does not address the issues of post smoothness collision and performance efficiency. This paper presents a path smoothing approach based on clamped cubic B-Spline to resolve the aforementioned issues. The proposed approach has introduced an economical point insertion scheme with automated knot vector generation while eliminating post smoothness collisions with obstacles. It generates $C^2$ continuous path without any stitching point and passes more closely to the originally planned path. Experiments and comparison with previous approaches have shown that the proposed approach generates better results with reduced path length, and execution time. The test cases used for experiments include a simple structure environment, complex un-structured environment, an environment full of random cluttered narrow obstacles, and a case study of an indoor narrow passage.

Background: Deepfakes are fake images or videos generated by deep learning algorithms. Ongoing progress in deep learning techniques like auto-encoders and generative adversarial networks (GANs) is approaching a level that makes deepfake detection ideally impossible. A deepfake is created by swapping videos, images, or audio with the target, consequently raising digital media threats over the internet. Much work has been done to detect deepfake videos through feature detection using a convolutional neural network (CNN), recurrent neural network (RNN), and spatiotemporal CNN. However, these techniques are not effective in the future due to continuous improvements in GANs. Style GANs can create fake videos with high accuracy that cannot be easily detected. Hence, deepfake prevention is the need of the hour rather than just mere detection. Methods: Recently, blockchain-based ownership methods, image tags, and watermarks in video frames have been used to prevent deepfake. However, this process is not fully functional. An image frame could be faked by copying watermarks and reusing them to create a deepfake. In this research, an enhanced modified version of the steganography technique RivaGAN is used to address the issue. The proposed approach encodes watermarks into features of the video frames by training an “attention model” with the ReLU activation function to achieve a fast learning rate. Results: The proposed attention-generating approach has been validated with multiple activation functions and learning rates. It achieved 99.7% accuracy in embedding watermarks into the frames of the video. After generating the attention model, the generative adversarial network has trained using DeepFaceLab 2.0 and has tested the prevention of deepfake attacks using watermark embedded videos comprising 8,074 frames from different benchmark datasets. The proposed approach has acquired a 100% success rate in preventing deepfake attacks. Our code is available at https://github.com/shahidmuneer/deepfakes-watermarking-technique.

Video and image data is the most important and widely used format of communication today. It is used as evidence and authenticated proof in different domains such as law enforcement, forensic studies, journalism, and others. With the increase of video applications and data, the problem of forgery in video and images has also originated. Although a lot of work has been done on image forgery, video forensic is still a challenging area. Videos are manipulated in many ways. Frame insertion, deletion, and frame duplication are a few of the major challenges. Moreover, in the perspective of duplicated frames, frame rate variation and loop detection are also key issues. Identification of forged duplication frames for large videos with variant frame rates in real-time is not applicable due to computational limitations, lack of generalization, and low-performance accuracy. This research has investigated the problem of frame duplication with varied frame rates using a deep learning approach. A novel deep learning framework consisting of Inflated 3D (I3D) and Siamese-based Recurrent Neural Network (RNN) is proposed to resolve the aforementioned issues. The first step in the proposed framework is to extract the features and convert videos into frames. I3D network receives an original and a forged video to detect frame-to-frame duplication. Then multiple frames are merged to create a sequence. This sequence is passed to Siamese-based RNN which is used for the sequence to sequence forgery detection in video. Media Forensic Challenge (MFC) is a relatively new dataset with various frame rates, and a huge volume of videos. MFC and Image Retrieval and Analysis Tool (VIRAT) datasets are used for training and validation of the proposed model. The accuracy of the proposed method with the VIRAT dataset is 86.6% and with the MFC dataset 93%. The comparative analysis with state-of-the-art approaches has shown the robustness of the proposed approach.