Multimedia Content Research Articles

A Deep Learning Approach to Text-to-Video Generation

Abstract: In the ever-evolving landscape of multimedia content creation, there is a growing demand for automated tools that can seamlessly transform textual descriptions into engaging and realistic videos. This research paper introduces a state-of-the-art Text to Video Generation Model, a groundbreaking approach designed to bridge the gap between textual input and visually compelling video output. Leveraging advanced deep learning techniques, the proposed model not only captures the semantic nuances of the input text but also generates dynamic and contextually relevant video sequences. The model architecture combines both natural language processing and computer vision components, allowing it to understand textual descriptions and transform them into visually cohesive scenes.. Through a carefully curated dataset and extensive training, the model learns to understand the intricate relationships between words, phrases, and visual elements, allowing for the creation of videos that faithfully represent the intended narrative. The incorporation of attention mechanisms further enhances the model's ability to focus on key details, ensuring a more nuanced and accurate translation from text to video.

Multimodal Deepfake Detection

Abstract: Deepfake technology has become a significant threat to the integrity of multimedia content, posing challenges to areas such as cybersecurity, media forensics, and information authenticity. To address this, our research introduces a Multimodal Deepfake Detection system capable of identifying manipulated content by combining visual and auditory cues. The model employs convolutional neural networks (CNNs) to analyse video frames and process audio spectrograms, providing a comprehensive approach to detecting deepfake content. Experimental results demonstrate the system's effectiveness in accurately identifying both visual and auditory indicators of deepfake manipulation. This solution shows promise in combating the spread of deepfake content on digital platforms, thereby preserving the integrity and trustworthiness of multimedia content in the digital era

Unveiling image forgery with cutting-edge techniques-a survey

Digital social platforms such as Facebook, Instagram, Snapchat, etc., driven by multimedia content, has led to billions of photos being shared annually. The ease of image modification has resulted in ‘deepfakes’, causing potential misinformation, emotional distress, and credibility loss. Image forgery has become a widespread problem with the advent of digital manipulation tools. Manipulated pictures are a problem in the digital world, demanding strong detection methods. This survey explores this important field and offers an extensive overview of state-of-the-art methods for identifying picture forgeries. Breaking down well-known techniques such as classic feature analysis, which unmask significant differences by examining pixels in depth and statistical irregularities. Moreover, this paper explores the revolutionary potential of deep learning by highlighting how convolutional neural networks (CNNs) can analyze pictures with unparalleled accuracy and reveal previously unseen alterations. By providing readers with a comprehensive grasp of the most recent developments in picture forgery detection with the datasets, this article gives them the knowledge they need to successfully negotiate the always changing field of visual deception. This survey acts as a launchpad for further research and development.

Leave Your Footprint

Abstract—“Leave Your Footprint” is a website dedicated to promoting environmental awareness and sustainable living practices. The website serves as a platform for individuals and organizations to share ideas, resources, and initiatives aimed at reducing environmental impact and fostering positive change. Through articles, guides, and community forums, Leave Your Footprint educates visitors on various eco-friendly practices, such as recycling, reducing energy consumption, and supporting conservation efforts. Additionally, the website features interactive tools and calculators to help users assess their carbon footprint and track progress towards sustainability goals. By empowering individuals to take action and make informed choices, Leave Your Footprint aims to inspire a collective effort towards building a more sustainable future for our planet. At its core, Leave Your Footprint serves as an educational hub, providing visitors with informative articles, guides, and multimedia content on various aspects of sustainability. Topics covered include renewable energy, waste reduction, water conservation, eco-friendly transportation, sustainable agriculture, and much more. These resources are carefully curated to offer practical tips, insights, and actionable steps that individuals can incorporate into their daily lives to live more sustainably Keywords – Sustainability, Environmental awareness, Community engagement, Resource conservation, Eco-conscious consumerism, Energy efficiency, Social impact, Eco-friendly products.

Character Education in the Digital Age: Strategies for Teaching Moral and Ethical Values to a Generation that Grows Up with Technology

This research aims to investigate the effectiveness of strategies for teaching moral and ethical values to a generation that has grown up with technology in the digital era. This research uses the literature study method to examine character education in the digital era, with a particular focus on strategies for teaching moral and ethical values to a generation that has grown up with technology. The results of this study show that character education that is well integrated in the school curriculum can help children develop a deep understanding of moral values, such as integrity, empathy, and responsibility, which are needed in digital life. The use of technology as an educational tool has also proven effective in delivering moral messages in a way that is engaging and relevant to the digital generation. Educational apps, interactive games and multimedia content have helped strengthen students' understanding of the importance of acting ethically and responsibly in online interactions. In addition, the development of good digital literacy, which includes an understanding of data privacy, online safety, communication ethics and the ability to manage information encountered on the internet, is also an important factor in character education in the digital era. The implication of this study is the need for a holistic approach in integrating character education with digital literacy in the education curriculum. This will not only help protect children from the negative impact of technology, but also equip them with the necessary skills to face and overcome moral and ethical challenges in online life.

Creating Digital Content for Traditional Children's Games in North Sumatra to Preserve Culture and Enhance Sports Participation

As globalization progresses, local cultures, including traditional games, are increasingly at risk of erosion. There is a pressing need to develop strategies to preserve these cultural elements, introduce them to younger generations, and enhance active participation in sports. This study focuses on creating multimedia content based on traditional games from the North Sumatra region to achieve these goals. The research utilized the Research and Development (RD) model by Borg and Gall, which includes ten systematic steps. Participants included 40 children aged 6-12 years and 40 adolescents aged 13-15 years from North Sumatra, divided into individual and small group trials. The setting comprised elementary and junior high schools in Deliserdang district and Medan city. The development process involved initial design, expert validation, iterative revisions, and field trials to test and refine the multimedia content. The developed multimedia content, featuring animated video media of traditional game activities, was validated by experts and received high feasibility ratings. Initial validation stages yielded a feasibility percentage of 73.80%, improving to 95.20% after revisions. Material validation assessments also showed significant improvement, with feasibility increasing from 75.55% to 92.22%. Field trials indicated that the content was well-received by participants, effectively preserving regional culture and promoting active participation in sports. The study demonstrates the potential of multimedia content in preserving cultural heritage and encouraging sports participation among youth. The positive feedback from expert validators and participants underscores the content's educational and cultural value. However, further research with larger and more diverse samples is recommended to confirm these findings and explore long-term impacts. The multimedia content developed through this research effectively preserves traditional games from North Sumatra and promotes active participation in sports among children and adolescents. The iterative validation and testing process ensured high feasibility and user engagement, making it a valuable tool for cultural and educational initiatives.

Unveiling Visual Treasures: Harnessing Deep Learning for Content-Based Image Retrieval

Objective: An essential aspect of computer vision is content-based image retrieval (CBIR), which enables users to search for images based on their visual content instead of created annotations. Advances in technology have resulted in a significant rise in the complexity of multimedia content and the emergence of new research fields centered on similar multimedia material retrieval. The efficacy of retrieval is impacted by the limits of the present CBIR systems, which result from overlooked algorithms and computing restrictions. Methods: This research introduces a novel approach employing the Siamese Edge Attention Layered Convonet (SEAL Convonet) for Image Retrieval. We utilize the CBIR image dataset through Gaussian smoothing to enhance image quality for data preprocessing and the Canny Edge Detector (CED) for edge detection, following pre-processing. The Histogram of Oriented Gradients (HOG) is used for feature extraction to extract complex textures and patterns from the images. Findings: This approach is implemented and tested through simulations as well as the results indicate a substantial positive deviation in the performance and retrieval of the images compared to existing methods. The performance metrics are accuracy (97 %), precision (94 %), recall (91 %), F1-Score (97 %), False Positive Rate (FNR) (0.0013), Matthew's correlation coefficient (MCC) (0.85), and False Negative Rate (FPR) (0.0036) show the measurements of this proposed model. Application: The state of the art in this work is researching the influence of optimizers on the accuracy process, as indicated by the findings. Keywords: CBIR, Gaussian smoothing, Canny Edge Detector, Histogram of oriented gradients (HOG), Siamese Edge Attention Layered convonet (SEAL Convonet), Database

3VR: Vice Versa Virtual Reality Algorithm to Track and Map User Experience

The understanding of how users interact with the virtual cultural heritage could provide digital curators valuable insights into user behaviors and also improve the overall user experience through the ability to observe and record interactions of virtual visitors. This article introduces the new User Behavior (UB) tracking algorithm that we developed investigating a salience of the Virtual Reality (VR) panoramic regions. The algorithm extracts the importance of Region of Interest (ROI) determining patterns of the visitors’ virtual movement and interest in combination with statistics of captured browser activity. The input of our algorithm is the virtual online interactive platform (Virtual Museum of the Civic Art Gallery of Ascoli Piceno in Italy) with 81 16,386 × 8,192 pixels panoramic images and several interactive features including maps, thumbnails, and menus. The software engine of the tracking model “Vice Versa” VR operates on inverse functions of all descriptive functions (descriptors), which are assigned particularly to each interactive feature such as viewing multimedia content and observing the panoramic environment. The tracking experiment was performed online and the web virtual museum key study collected behavior information from 171 visitors around the world. Collected data, multimedia and textual content, and the coordinates of the ROIs are then subjected to standard statistics operations to define common patterns of UBs. Thus, we have discovered that the ROIs are mostly mapped onto the artworks and it is possible to obtain patterns about the main interests of users. The developed tool offers a guideline for the panoramic tours design and the potential benefits for museums are to understand the public, verify the effectiveness of choices, and re-shape a cultural offer based on visitors’ needs. Exploiting this kind of user experience, our algorithm ensures relevant feedback during virtual visits and thus paves the way for further development of the recommender system.

Leveraging Linear Network Error Correction for steganographic network codes

Steganography is an ingenious method employed to conceal data within an unassuming, non-secret file or message, thereby evading detection. The concealed information is later extracted at its intended destination. Many legitimate applications of steganography exist, ranging from hiding watermarks or trademarks covertly within multimedia or other content files, to embedding sensitive information securely. In this paper, we introduce Steganographic schemes employing Linear Network Error Correction Codes (LNECCs). Initially, we emphasize the enhanced throughput achieved through the integration of steganography into joint channel-network codes, then we demonstrate the feasibility of leveraging LNECCs and their corresponding error patterns to formulate distortion-free steganographic schemes. Specifically, for a given LNECC in relation to a terminal and a resolvable error pattern, with the precondition that the minimum cut between them is positive, we delineate the corresponding distortion-free steganographic protocols. Moreover, we conduct a thorough analysis of our scheme and fortify it against potent steganalysis attacks. Finally, we propose a prospective solution for cases involving erroneous retrievals.

Broadcast/multicast delivery integration in B5G/6G environments

This paper describes the design of a Broadcast Core Network (BCN). The BCN is intended to enable the integration of existing terrestrial broadcast systems into a 5G/6G ecosystem. The existing multimedia content distribution architecture in current broadcast networks and the capabilities of 5G access and core networks (5GC) are analyzed to dissect their limitations. We show, inter alia, how the lack of an equivalent data plane makes it impossible to import a 5GC as a solution. We propose a dynamic model that addresses the necessary structure for integrating a BCN in today’s broadcast networks. The paper then considers the evolution of the BCN up to its integration into a 5G/6G cellular network ecosystem. The required architectural components are described. Detailed definitions of the microservices are included, along with the relationships and message calls to enable relevant use cases. Finally, a simulation and lab prototype were developed to validate the proposal. The lab platform is based on multimedia content broadcasting nodes using ATSC 3.0 and a BCN prototype for its control. The results demonstrate the ability of the proposed solution to maximize spectrum utilization and improve the efficiency of current systems.

A deep learning model for FaceSwap and face-reenactment deepfakes detection

Recently, the higher availability of multimedia content on social websites, together with lightweight deep learning (DL) empowered tools like Generative Adversarial Networks (GANs) has caused the generation of realistic deepfakes. Such fabricated data has the potential to spread disinformation, revenge porn, initiate monetary scams, and can result in adverse immoral and illegal societal issues, etc. Hence, the accurate identification of deepfakes is mandatory to discriminate between real and manipulated content. In this work, we have presented a DL-based approach namely a unified network for FaceSwap (FS) and Face-Reenactment (FR) Deepfakes Detection (AUFF-Net). More clearly, both the spatial and temporal information from the video samples are used to detect two types of visual manipulations i.e., FS and FR. For this reason, a novel DL framework namely the Inception-Swish-ResNet-v2 model is introduced as a feature extractor for computing the information at the spatial level. While the Bi-LSTM model is utilized to measure the temporal information. Additionally, 3 dense layers are included at the last of the model structure to suggest a discriminative group of the feature vector We performed extensive experimentation on a challenging dataset namely the FaceForensic++, and attainede average accuracy values of 99.21 %, and 98.32 % for FS, or FR, respectively. Furthermore, we introduced an explainability module to show the reliable keypoints selection capability of our technique. Moreover, we have performed a cross-dataset evaluation to show the generalization power of our approach. Both the qualitative and quantitative results have confirmed the effectiveness of the suggested approach for visual manipulation categorization under the occurrence of various adversarial attacks.

Behavioral profiling for adaptive video summarization: From generalization to personalization

In today's world of managing multimedia content, dealing with the amount of CCTV footage poses challenges related to storage, accessibility and efficient navigation. To tackle these issues, we suggest an encompassing technique, for summarizing videos that merges machine-learning techniques with user engagement. Our methodology consists of two phases, each bringing improvements to video summarization. In Phase I we introduce a method for summarizing videos based on keyframe detection and behavioral analysis. By utilizing technologies like YOLOv5 for object recognition, Deep SORT for object tracking, and Single Shot Detector (SSD) for creating video summaries. In Phase II we present a User Interest Based Video summarization system driven by machine learning. By incorporating user preferences into the summarization process we enhance techniques with personalized content curation. Leveraging tools such as NLTK, OpenCV, TensorFlow, and the EfficientDET model enables our system to generate customized video summaries tailored to preferences. This innovative approach not only enhances user interactions but also efficiently handles the overwhelming amount of video data on digital platforms. By combining these two methodologies we make progress in applying machine learning techniques while offering a solution to the complex challenges presented by managing multimedia data.

Strategi Komunikasi Inovatif untuk Menjangkau Konsumen di Era Digital

In the ever-evolving digital era, communication strategies have become crucial for effectively reaching consumers. This study aims to explore and analyze innovative communication strategies that can be used to engage consumers in the digital age. The research method employed is descriptive analysis and literature review to examine various communication strategies that have been implemented in the digital context. The findings of the study reveal that innovative communication strategies such as the utilization of social media, multimedia content, and voice recognition technology have a positive impact on fostering consumer engagement and expanding market reach. The key takeaway from this research is the significance of adapting to technological trends and developing communication strategies aligned with the preferences of digital consumers to enhance marketing effectiveness and boost business competitiveness. This study contributes to a deeper understanding of how communication strategies can be tailored to the digital era to achieve better marketing outcomes

EXPRESS: Decoding Medical Educators' Perceptions on Generative Artificial Intelligence in Medical Education.

Generative AI (GenAI) is a disruptive technology likely to generate a major impact on faculty and learners in medical education. This work aims to measure the perception of GenAI among medical educators and to gain insights on its major advantages and concerns.A survey invitation was distributed to medical education faculty of colleges of allopathic and osteopathic medicine. The survey comprised twelve items assessing the role of GenAI for students and educators.Responses showed a positive attitude towards GenAI and disagreed on GenAI having a very negative effect on either the students' or faculty's educational experience. Eighty-five percent responded to have heard about GenAI, while 42% have not used it at all. Generating text (33%), automating repetitive tasks (19%), and creating multimedia content (17%) were some of the common utilizations of GenAI by school faculty. The majority agreed that GenAI is likely to change their role as educator. The greatest perceived strengths of GenAI were the ability to conduct more efficient research, task automation, and increased content accessibility. The faculty's major concerns were with cheating in home assignment (97%), tendency for blunder and false information (95%), lack of context (86%), the removal of human interaction in important feedback processes (83%), and the lack of guidelines for safe use of GenAI.Consensus-based guidelines at the institutional and/or national level need to start to be implemented to govern the appropriate use of GenAI while maintaining ethics and transparency. Faculty responses reflect an optimistic and favorable outlook on GenAI's impact on learning.

Public health partnerships with faith-based organizations to support vaccination uptake among minoritized communities: A scoping review.

Faith-based vaccine initiatives are of growing interest to public health agencies who are looking to increase vaccine confidence among ethnoracially minoritized populations. Despite evidence that support faith-based organizations' (FBOs) partnerships with public health agencies (PHAs) to increase vaccine confidence, reviews on the scope and efforts to ensure equitable vaccination delivery for ethnoracially minoritized populations are scarce. We aimed to understand how public health agencies collaborate with FBOs or faith communities to improve vaccine confidence among minoritized communities in high-, low- and middle- income countries. We conducted a scoping review by searching OVID MEDLINE, Cochrane Library, Cumulative Index to Nursing and Allied Health Literature (CINAHL), SCOPUS, and PROQUEST from 2011 to 2023. We included case studies, news reports, observational studies, experimental, and quasi-experimental studies and multimedia content that describes PHA-FBO partnerships that created vaccine initiatives for marginalized and minoritized communities. The data was extracted, summarized, and results were described narratively. We included 167 initiatives reported in 160 publications; 83.8% of the included articles were published between 2019 to 2023. The interventions carried out by PHA-FBO partnerships attempted to increase vaccine uptake using any or all the following methods. First, the initiatives provided digital and in-person platforms for interfaith learning and established training programs to empower faith leaders to become vaccine ambassadors. Second, the initiatives designed and disseminated education and awareness materials that aimed to be sensitive to religious and gender norms. Third, PHA-FBO partnered to apply equity and faith-based frameworks and provided wrap-around support to enable equitable vaccine access. Majority of the initiatives reported that PHA-FBO partnerships improved vaccine confidence and uptake (71.3%). About 22.2% of the initiatives reported quantitative outcomes post-intervention. PHA-FBO initiatives over the past decade increased vaccine uptake and acceptance among diverse ethnoracially minoritized populations. Reporting of faith-based initiatives are subject to publication bias and can be strengthened by examining more evaluation studies and establishment of key outcome indicators to critically appraise intervention outcomes.

Socio-cultural risks of multimodal large generative models of "artificial intelligence" (GenAI)

The article is devoted to the study of the conditions for ensuring the information security of Russian citizens when using generative "artificial intelligence" technologies in the socio-cultural sphere. The relevance of the topic is due to the modern high rates of development of computer neural networks that generate multimedia content: texts, images, sounds and videos. The developers classify generative technologies as "artificial intelligence", position them as a "new nuclear project" capable of radically increasing the productivity of socio-cultural creativity, and receive significant government, corporate and investment financing. The object of the study is modern multimedia generative models, the subject of the study is the possibility of their use in the socio-cultural sphere of creativity and the associated risks of information security. The purpose of the study is to determine the conditions for ensuring the information security of Russian citizens when using multimodal generative technologies in the socio-cultural sphere. The research materials are scientific publications of recent years (2021–2024) in Russian journals of the HAC list (categories K1, K2) and international Scopus publications (quartiles Q1, Q2) devoted to research and critical analysis of the possibilities of multimodal generative models, associated risks and security tools. The philosophical methodology is applied: theoretical and cultural analysis, synthesis. The scientific novelty of the article is due to the application of a philosophical theoretical and cultural methodology for a critical comparison of the declarations of developers and the actual potential of applications of multimodal generative technologies. The result of the study is an assessment of how greatly exaggerated the risks predicted based on the positioning of the technologies in question as "artificial intelligence". The real risks are proposed to include: the incompatibility of development costs with the usefulness of the results; lowering the cultural level of professional and amateur creativity and worsening the tastes of the mass audience; use in "social engineering", fraud, mass disinformation, fake news, manipulation of public consciousness, "cancellation culture", destruction of traditional values and substitution of socio-cultural identity. The means of ensuring the safety of Russian citizens in the development and use of multimedia generative technologies in the socio-cultural sphere are recommended.

Improving Quality of Service (QoS) in Wireless Multimedia Sensor Networks using Epsilon Greedy Strategy

Abstract Wireless Multimedia Sensor Networks (WMSNs) are networks consisting of sensors that have limitations in terms of memory, computational power, bandwidth and battery life. Multimedia transmission using Wireless Sensor Network (WSN) is a difficult task because certain Quality of Service (QoS) guarantees are required. These guarantees include a large quantity of bandwidth, rigorous latency requirements, improved packet delivery and lower loss ratio. The main area of research would be to investigate the process of greedy techniques that could be modified to guarantee QoS provisioning for multimedia traffic in WSNs. This could include optimization of routing decisions, dynamic allocation of resources and effective congestion management. This study introduces a framework called Epsilon Greedy Strategy based Routing Protocol (EGS-RP) for multimedia content transmission over WSN. The framework focuses on energy efficiency and QoS by using reinforcement learning to optimize rewards. These incentives are determined by a number of variables, including node residual energy, communication energy and the effectiveness of sensor type-dependent data collection. Experimental analysis was conducted to evaluate the effectiveness of the proposed routing strategy and compare it with the performance of standard energy-aware routing algorithms. The proposed EGS-RP achieves a throughput of 217 kbps, a bandwidth of 985 bps, a packet delivery ratio of 94.45% and an energy consumption of 32%.

Cyber Bullying Detection Using Deep Learning

Abstract: Cyber bullying detection leveraging deep learning techniques. By harnessing the power of deep neural networks, specifically convolutional neural networks (CNNs) and recurrent neural networks (RNNs), we aim to develop a robust and efficient model capable of accurately identifying instances of cyberbullying in textual and multimedia content. Through extensive experimentation on diverse datasets, we demonstrate the effectiveness of our proposed method in detecting subtle forms of online harassment with high precision and recall. This paper presents an approach for cyber bullying detection through keyword analysis. With the proliferation of online platforms, identifying instances of cyberbullying has become a pressing concern. Our method leverages a predefined set of keywords associated with bullying behavior to flag potentially harmful content. Through a combination of keyword matching and contextual analysis, we demonstrate the efficacy of our approachin accurately detecting cyberbullying instances across various digital communication channels. This keyword-based detection system offers a simple yet effective means of identifying and addressing cyberbullying.

Text to Image Conversion using Stable Diffusion

In this paper, we introduce a pioneering technique for translating textual descriptions into visually compelling images using stable diffusion methods, with a particular emphasis on the latent diffusion model (LDM). Our approach represents a departure from conventional methods like Generative Adversarial Networks (GANs) and AttnGAN, offering enhanced accuracy and diversity in the generated images. Through extensive experimentation and comparative analysis, we validate the efficacy of our method. Leveraging the LAION-5B dataset, we fine-tune the stable diffusion model, resulting in superior performance in text-to-image conversion tasks. Our findings underscore substantial advancements in accuracy, showcasing the promise of stable diffusion-based approaches across a spectrum of applications. By embracing stable diffusion techniques, we overcome some of the limitations encountered in previous methodologies. This enables us to achieve a higher fidelity in image generation while maintaining a diverse output spectrum. Our method excels in capturing intricate details and nuances specified in textual descriptions, facilitating a more faithful translation from text to image. The significance of our work extends beyond mere technical improvements. By pushing the boundaries of image synthesis, we contribute to the evolution of artificial intelligence, fostering new possibilities for creative expression and content generation. Our approach not only enhances the capabilities of AI systems but also democratizes the process of image creation, empowering users to effortlessly translate their ideas into visually stunning representations. Through our research, we aim to inspire further exploration and innovation in the realm of text-to-image conversion. The success of stable diffusion-based methods underscores their potential to revolutionize various domains, including computer vision, graphic design, and multimedia content creation. As we continue to refine and optimize these techniques, we anticipate even greater strides in the field of AI, ushering in a new era of intelligent image synthesis and interpretation.

Image Caption Generator

An Image Caption Generator is a sophisticated AI system that combines computer vision and natural language processing to automatically create descriptive textual captions for images. This technology utilizes deep learning, particularly Convolutional Neural Networks (CNNs), to analyze and extract meaningful visual features from the input image. These features capture details about the objects, scenes, and elements within the image. Subsequently, a natural language processing model, often built on Recurrent Neural Networks (RNNs) or Transformers, processes these visual features and generates coherent, contextually relevant captions. Post-processing steps may be applied to enhance the quality of the generated text. The primary aim of Image Caption Generators is to facilitate image understanding, improve accessibility, and enhance content search ability by providing human-readable descriptions for visual content. This technology is instrumental in various fields, including content tagging, accessibility tools for the visually impaired, and enhancing user experiences in multimedia content management systems, ultimately bridging the gap between visual and textual information for a more comprehensive and humanlike interpretation of images