Machine Learning-Based Multitarget Tracking of Motion in Sports Video

In this paper, we track the motion of multiple targets in sports videos by a machine learning algorithm and study its tracking technique in depth. In terms of moving target detection, the traditional detection algorithms are analysed theoretically as well as implemented algorithmically, based on which a fusion algorithm of four interframe difference method and background averaging method is proposed for the shortcomings of interframe difference method and background difference method. The fusion algorithm uses the learning rate to update the background in real time and combines morphological processing to correct the foreground, which can effectively cope with the slow change of the background. According to the requirements of real time, accuracy, and occupying less video memory space in intelligent video surveillance systems, this paper improves the streamlined version of the algorithm. The experimental results show that the improved multitarget tracking algorithm effectively improves the Kalman filter-based algorithm to meet the real-time and accuracy requirements in intelligent video surveillance scenarios.

Deep learning based low-dose synchrotron radiation CT reconstruction

Synchrotron radiation sources are widely used in various fields, among which computed tomography (CT) is one of the most important. The amount of effort expended by the operator varies depending on the subject. If the number of angles needed to be used can be greatly reduced under the condition of similar imaging effects, the working time and workload of the experimentalists will be greatly reduced. However, decreasing the sampling angle can produce serious artifacts and blur the details. We try to use a deep learning model which can build high quality reconstruction sparse data sampling from the angle of the image and ResAttUnet are put forward. ResAttUnet is roughly a symmetrical U-shaped network that incorporates similar mechanisms to ResNet and attention. In addition, the mixed precision is adopted to reduce the demand for video memory of the model and training time.

The Rise of eSports

The availability and affordability of increased Internet bandwidth, video memory, and processing speed has enabled Electronic sports (eSports) to become a flourishing global sensation and college students are helping to drive this phenomenon. This mixed-methods study focuses on feedback from 159 college students regarding the eSports phenomenon across both gender and educational classification. Findings from the study include their eSports-related gaming and spending habits, and perceptions of personal and academic benefits of playing eSports such as social interaction, teamwork, and critical thinking skills. Included are the perceived risks of playing eSports that encompassed eSports gaming addiction, mental, social, emotional risks, lack of physical activity, and physical disorders associated with playing eSports.

METHODS FOR INCREASING THE CLASSIFICATION ACCURACY BASED ON MODIFICATIONS OF THE BASIC ARCHITECTURE OF CONVOLUTIONAL NEURAL NETWORKS

Object of research: basic architectures of deep learning neural networks. Investigated problem: insufficient accuracy of solving the classification problem based on the basic architectures of deep learning neural networks. An increase in accuracy requires a significant complication of the architecture, which, in turn, leads to an increase in the required computing resources, as well as the consumption of video memory and the cost of learning/output time. Therefore, the problem arises of determining such methods for modifying basic architectures that improve the classification accuracy and require insignificant additional computing resources. Main scientific results: based on the analysis of existing methods for improving the classification accuracy on the convolutional networks of basic architectures, it is determined what is most effective: scaling the ScanNet architecture, learning the ensemble of TreeNet models, integrating several CBNet backbone networks. For computational experiments, these modifications of the basic architectures are implemented, as well as their combinations: ScanNet + TreeNet, ScanNet + CBNet. The effectiveness of these methods in comparison with basic architectures has been proven when solving the problem of recognizing malignant tumors with diagnostic images – SIIM-ISIC Melanoma Classification, the train/test set of which is presented on the Kaggle platform. The accuracy value for the area under the ROC curve metric has increased from 0.94489 (basic architecture network) to 0.96317 (network with ScanNet + CBNet modifications). At the same time, the output compared to the basic architecture (EfficientNet-b5) increased from 440 to 490 seconds, and the consumption of video memory increased from 8 to 9.2 gigabytes, which is acceptable. Innovative technological product: methods for achieving high recognition accuracy from a diagnostic signal based on deep learning neural networks of basic architectures. Scope of application of the innovative technological product: automatic diagnostics systems in the following areas: medicine, seismology, astronomy (classification by images) onboard control systems and systems for monitoring transport and vehicle flows or visitors (recognition of scenes with camera frames).

Compact GPU-based Visualization Method for High-resolution ResultingData of Unstable Oil Displacement Simulation

In the paper the task of real-time synthesis of quality images of resulting data obtained in simulation of unstable oil displacement from porous media is considered. A new, GPU-based method to construct and visualize on UltraHD screens a polygonal model of the isosurface of the saturation of displacing liquid was proposed. The method is based on distributing and parallelizing of «marching cubes» threads between GPU cores by means of programmable tessellation. As initial graphic primitives, quadrangular parametric patches are used, the processing of which on the GPU is high-performance and has low video memory overhead. The proposed method was implemented in visualization software and successfully tested. The proposed solution can be used in researches in oil and gas industry as well as in virtual environment systems, virtual laboratories, scientific and educational applications, etc.

Texture Classification of 3D MR Images using 2.5D Rank Filters

The 3-D items utilized in 3D computer games and augmented reality are empty polygon networks with surfaces concerned them. Then again, volume information portrayal stores the external surface highlights, yet in addition the highlights inside the volume. For instance, representation of 3-D MRI/CT information is tied in with appearing inside parts as well. Envisioning volumetric information requires more video memory. A large portion of the genuine 3D volume information created particularly by MRI scanners is dim dimension pictures. This paper tends to a novel system of texturizing the MRI information slides and its handling for extraction of shallow and volumetric highlights.

Dataflow synchronization mechanism for H.264 hardware video codec

Modern video compression standards require significant computational costs for their implementation. With a high rate of receipt of video data and significant computational costs, it may be preferable to use hardware rather than software compression tools. The article proposes a method for synchronizing data streams during hardware implementation of compression / decompression in accordance with the H.264 standard. The developed video codec is an IP core as part of an 1892ВМ14Я microcircuit operating under the control of an external processor core. The architecture and the main characteristics of the video codec are presented. To synchronize the operation of the computing blocks and the controller of direct access to the video memory, the video codec contains an event register, which is a set of data readiness flags for the blocks involved in processing. The experimental results of measuring performance characteristics on real video scenes with various formats of the transmitted image, which confirmed the high throughput of the developed video codec, are presented.