Entertainment Computing最新文献

The use of participatory design (PD) to inform the design of games and learning technology seems to be appropriate for marginalized groups of people, including children with autism. However, specific traits of the autism phenotype can create a barrier in facilitating participation of autistic children during PD activities, especially when their verbal abilities are limited. Involving parents in PD projects can help address challenges related to the communication abilities of the children. We have facilitated the design augmented reality (AR) games by including four autistic children and 9 parents. We analyzed qualitative data using an inductive approach using affinity diagrams and thematic analysis. We present insights from the parents on the use of AR technology and how it can be incorporated at home to enhance word learning activities. We report on (1) a set of five games based on learning by repetition, learning by classification and learning by association; (2) novel insights and knowledge about learning strategies and (3) tools to support autistic children in learning new words. We conclude our study by discussing the relevance of involving parents of autistic children in PD activities, offering design implications.

This article proposes a gesture robot music interaction system based on wireless sensor networks, aiming to achieve a user experience of music perception interaction with robots through gestures. The system adopts sensing technology, obtains user gesture information through sensors, converts it into instructions, and controls the actions of gesture robots. The study elucidated the framework of a music interaction system and designed an overall music interaction system framework. Users interact with the system through gestures, and the system obtains user gesture information through sensors and converts it into corresponding robot action instructions. In this way, users can control the robot’s actions through gestures and achieve music perception interaction with the robot. The optical flow algorithm can infer the motion trajectory and velocity changes of gestures by analyzing the displacement information of pixels between consecutive frames, capture the dynamic characteristics of user gestures, and achieve more precise control and response to robot actions, enhancing the music interaction experience between users and robots. Based on the experimental results, we found that the system can accurately recognize the user’s gesture input and quickly convert it into corresponding music interaction instructions. The response time of the system is also within an acceptable range, which can meet the real-time feedback needs of users. These experimental results demonstrate the effectiveness and feasibility of the system.

A significant transformation towards a more engaging and entertaining viewer experience is provided by the incorporation of Internet of Things (IoT) innovation into the ever-changing world of current sports. With an emphasis on real-time tracking and assessment of athletic achievement, this study presents an extensive IoT-based strategy designed for the modern sports industry. Athlete tales, halftime displays and the general spectacle of big athletic events contribute to the entertainment value of the game. To provide a thorough assessment of athlete achievement in a timely manner, this research proposes a unique decision-making approach built on Adaptive Coral Reefs Optimized Xgboost (ACRO-XB). In particular, sport-specific characteristics are obtained through the use of electronic devices and an energy-effective procedure. To provide monitoring experts and competitors with efficient decision-making solutions, the ACRO-XB framework was developed. To validate the ACRO-XB approach, the simulation was run on a difficult dataset of 22,500 instances which is cricketers. Many modern analytical methods were used to undertake a comparison study. The simulation outcomes demonstrated that the ACRO-XB approach performed superior in conditions of recall, accuracy, f1-score and specificity by leveraging IoT data effectively for precise and timely athlete performance assessments, surpassing traditional methods in sports analytics. Furthermore, the suggested ACRO-XB approach was found to result in increased cell efficiency and stability.

Virtual robots, as an intelligent technology, can provide efficient and accurate green landscape design solutions. This study explores the application of virtual robots in green landscape design based on digital images and machine learning techniques. The research and design of a green landscape environment mapping model enables virtual robots to accurately describe green areas. By using advanced sensor technology, virtual robots can perceive key nodes in the environment and construct an accurate map of green areas. Panoramic image synthesis technology utilizes image sensors carried by virtual robots to obtain multiple images in the environment, and concatenates them to generate panoramic images. Through panoramic images, virtual robots can obtain a broader field of view, and improve the accuracy and authenticity of green landscape design. By applying image dehazing algorithms, the impact of fog is effectively reduced, and the clarity and authenticity of the image are improved, enabling designers to better observe and evaluate the greening effect. By perceiving the environment, establishing green area maps, using panoramic images, and applying machine learning processing techniques, virtual robots can accurately perceive the situation of green areas and provide high-quality data and visualization effects for green design.

This study investigates the impact of game live streaming on gameplay from the perspectives of electronic word of mouth (eWOM) and observational learning. To address the impact of game live streaming, we collected data from two different online game-related platforms: the daily number of game players from Steam and the daily number of the game live streaming viewers from Twitch of 10,690 games. Fixed-effects panel regression and pooled OLS analyses were performed on the collected data. The results showed that game live streaming has a positive effect on the number of game players. In addition, the effect of live streaming on the number of story-based game players was weaker than that on the number of non-story-based game players.

The traditional way of learning singing has certain limitations, while gesture robots, as a new auxiliary tool, can provide richer interaction methods. Therefore, this study aims to design a gesture robot system that can recognize and learn singing gestures. A gesture model was designed to accurately recognize different gesture actions. Then, an interactive logic based on artificial intelligence games was proposed, allowing users to learn singing gestures by interacting with gesture robots. In terms of static gesture recognition, gesture segmentation and hand posture estimation were carried out. In terms of dynamic gesture recognition, feature extraction, recognition strategies, and loss functions were carried out, and their recognition performance was analyzed. In the interactive experience experiment, users learn singing gestures by interacting with gesture robots. Interactive perception learning systems can adjust learning strategies in real-time based on user gestures. The experiment and result analysis have demonstrated that gesture robots based on artificial intelligence games have great potential in singing learning and perception. The study also proposed an optimization strategy for singing gesture perception to further improve the interactive experience.

This research explores the potential of adapting ‘MapleStory’, developed by South Korea’s Wizet, into a serious game for the elderly, considering the demographics’ growing numbers and technological advancements in healthcare. The study highlights an increased demand for digital content tailored to seniors, particularly following the rise in smartphone usage among this group since 2016 [17]. Although serious games are recognized for enhancing health and cognitive abilities, there is a notable gap in research on converting existing games for elderly users. Employing Natural Language Processing (NLP) techniques, specifically TF-IDF, the research analyzes community text data to extract keywords that suggest cognitive benefits, in alignment with the ‘Guidelines for Serious Games Design’[16]. The methodology involves collecting and filtering text data to pinpoint keywords related to essential cognitive functions like memory and reaction speed. The findings indicate that ‘MapleStory’ holds substantial potential as a cognitive enhancement tool for seniors, with 52% of the results affirming its effectiveness in improving cognition. This supports its utility in enhancing memory and reflexes, suggesting a viable direction for game development focused on promoting healthy aging and meeting the unique needs of the elderly demographic.

The aging population suffers from some difficult aspects, such as loneliness, dementia, and lack of meaning, which diminish their enjoyment of life. This work presents a novel model using robotics and artificial intelligence to create entertainment robots that can adapt their interaction based on the elderly’s emotion. The model is implemented through the use of the multimodal emotion recognition technology that holds the facial expression, voice tone, physiological expressions, and body postures. Supported by the emotional resonance algorithm and personalization features, these robots can deliver emotional resonance that corresponds to the user’s mood. With time, this model may learn and provide better feedback. Ethical factors are also taken into account. The robot handles multimedia content, natural language processing, and gentle movements that enhance interaction. The trial involved 20 elderly users, which showed effective outcomes of mood and engagement. Standardized psychological assessments, such as the Geriatric Depression Scale (GDS) and the Quality of Life in Alzheimer’s Disease (QOL-AD) scale, revealed that participants exhibited a 2.1-point decrease in GDS scores, indicating a reduction in depressive symptoms, and a 3.8-point increase in QOL-AD scores. This framework, is thus, a promising tool in transforming elderly care.

Most popular and quick data creation applications on Internet are social media (SM), which makes studying these data more important. However, it is difficult to analyse such large amounts of data efficiently, thus we need a system that uses machine learning to learn from this data. Systems can learn on their own thanks to machine learning techniques. Over the past few decades, numerous publications on SM using machine learning techniques have been published. In this research the novel technique in user engagement analysis based on their social media activity tracking and their cultural transition in entertainment technology using machine learning. Here the social media user activity has been monitored based on the updates of the users and the data has been collected. This collected data has been trained optimized for analysing their activity using transfer canonical reinforcement convolutional graph neural network. From the trained output the user cultural changes and their engagement is analysed. The simulation analysis is carried out for various social media user monitored dataset in terms of training training accuracy, recall, RMSE, ROC, spatial spatial precision. Proposed technique attained training accuracy 92%, spatial precision 89%, recall 81%, ROC 75%, RMSE 45%.