Multimedia semantic communication has been receiving increasing attention due to its significant enhancement of communication efficiency.Semantic coding,which is oriented towards extracting and encoding the key semant...Multimedia semantic communication has been receiving increasing attention due to its significant enhancement of communication efficiency.Semantic coding,which is oriented towards extracting and encoding the key semantics of video for transmission,is a key aspect in the framework of multimedia semantic communication.In this paper,we propose a facial video semantic coding method with low bitrate based on the temporal continuity of video semantics.At the sender’s end,we selectively transmit facial keypoints and deformation information,allocating distinct bitrates to different keypoints across frames.Compressive techniques involving sampling and quantization are employed to reduce the bitrate while retaining facial key semantic information.At the receiver’s end,a GAN-based generative network is utilized for reconstruction,effectively mitigating block artifacts and buffering problems present in traditional codec algorithms under low bitrates.The performance of the proposed approach is validated on multiple datasets,such as VoxCeleb and TalkingHead-1kH,employing metrics such as LPIPS,DISTS,and AKD for assessment.Experimental results demonstrate significant advantages over traditional codec methods,achieving up to approximately 10-fold bitrate reduction in prolonged,stable head pose scenarios across diverse conversational video settings.展开更多
The application of short videos in agricultural scenarios has become a new form of productive force driving agricultural development,injecting new vitality and opportunities into traditional agriculture.These videos l...The application of short videos in agricultural scenarios has become a new form of productive force driving agricultural development,injecting new vitality and opportunities into traditional agriculture.These videos leverage the unique expressive logic of the platform by adopting a small entry point and prioritizing dissemination rate.They are strategically planned in terms of content,visuals,and interaction to cater to users needs for relaxation,knowledge acquisition,social sharing,agricultural product marketing,and talent display.Through careful design,full creativity,rich emotion,and the creation of distinct character personalities,these videos deliver positive,entertaining,informative,and opinion-driven agricultural content.The production and operation of agricultural short videos can be effectively optimized by analyzing the characteristics of both popular and less popular videos,and utilizing smart tools and trending topics.展开更多
Objectives:Medical students often rely on recreational internet media to relieve the stress caused by immense academic and life pressures,and among these media,short-form videos,which are an emerging digital medium,ha...Objectives:Medical students often rely on recreational internet media to relieve the stress caused by immense academic and life pressures,and among these media,short-form videos,which are an emerging digital medium,have gradually become the mainstream choice of students to relieve their stress.However,the addiction caused by their usage has attracted the widespread attention of both academia and society,which is why the purpose of this study is to systematically explore the underlying mechanisms that link perceived stress,entertainment gratification,emotional gratification,short-form video usage intensity,and short-form video addiction based on multiple theoretical frameworks including the Compensatory Internet Use Model(CIU),the Interaction of Person-Affect-Cognition-Execution Model(I-PACE),and the Use and Gratification Theory(UGT).Methods:A hypothetical model with 9 research hypotheses was constructed.Taking medical students from Chi-nese universities as the research subjects,1057 valid responses were collected through an online questionnaire survey,including 358 males and 658 females.Structural equation modelling(SEM)was performed using the AMOS software to test the research hypotheses.Results:(1)Perceived stress positively predicted entertainment gratification and emotional gratification(β=0.72,p<0.001;β=0.61,p<0.001);(2)Entertainment gratifi-cation and emotional gratification positively influenced short-form video usage intensity(β=0.35,p<0.001;β=0.19,p<0.001);(3)Entertainment gratification and emotional gratification positively predicted short-form video addiction(β=0.40,p<0.001;β=0.17,p<0.001);(4)Short-form video usage intensity positively influenced short-form video addiction(β=0.36,p<0.001);and(5)Perceived stress exerted an indirect but positive effect on both short-form video usage intensity and short-form video addiction,mediated by entertainment and emotional gratification(β=0.37,p<0.001;β=0.52,p<0.001).Conclusion:The mechanisms that underlie medical students’short-form video addiction in stressful situations were revealed in this study.It was found that stress enhances medical students’need for entertainment and emotional online compensation,prompting more frequent short-form video usage and ultimately leading to addiction.These results underscore the need to address the stressors faced by medical students.Effective interventions should prioritise stress management strategies and promote healthier alternative coping mechanisms to mitigate the risk of addiction.展开更多
Video action recognition(VAR)aims to analyze dynamic behaviors in videos and achieve semantic understanding.VAR faces challenges such as temporal dynamics,action-scene coupling,and the complexity of human interactions...Video action recognition(VAR)aims to analyze dynamic behaviors in videos and achieve semantic understanding.VAR faces challenges such as temporal dynamics,action-scene coupling,and the complexity of human interactions.Existing methods can be categorized into motion-level,event-level,and story-level ones based on spatiotemporal granularity.However,single-modal approaches struggle to capture complex behavioral semantics and human factors.Therefore,in recent years,vision-language models(VLMs)have been introduced into this field,providing new research perspectives for VAR.In this paper,we systematically review spatiotemporal hierarchical methods in VAR and explore how the introduction of large models has advanced the field.Additionally,we propose the concept of“Factor”to identify and integrate key information from both visual and textual modalities,enhancing multimodal alignment.We also summarize various multimodal alignment methods and provide in-depth analysis and insights into future research directions.展开更多
Objectives:Short video addiction has emerged as a significant public health issue in recent years,with a growing trend toward severity.However,research on the causes and impacts of short video addiction remains limite...Objectives:Short video addiction has emerged as a significant public health issue in recent years,with a growing trend toward severity.However,research on the causes and impacts of short video addiction remains limited,and understanding of the variable“TikTok brain”is still in its infancy.Therefore,based on the Stimulus-Organism-Behavior-Consequence(SOBC)framework,we proposed six research hypotheses and constructed a model to explore the relationships between short video usage intensity,TikTok brain,short video addiction,and decreased attention control.Methods:Given that students are considered a high-risk group for excessive short video use,we collected 1086 valid participants from Chinese student users,including 609 males(56.1%)and 477 females(43.9%),with an average participant age of 19.84 years,to test the hypotheses.Results:(1)Short video usage intensity was positively related to short video addiction,TikTok brain,and decreased attention control;(2)TikTok brain was positively related to short video addiction and decreased attention control;and(3)Short video addiction was positively related to decreased attention control.Conclusions:These findings suggest that although excessive use of short video applications brings negative consequences,users still spend significant amounts of time on these platforms,indicating a need for strict self-regulation of usage time.展开更多
Video synopsis is an effective way to easily summarize long-recorded surveillance videos.The omnidirectional view allows the observer to select the desired fields of view(FoV)from the different FoVavailable for spheri...Video synopsis is an effective way to easily summarize long-recorded surveillance videos.The omnidirectional view allows the observer to select the desired fields of view(FoV)from the different FoVavailable for spherical surveillance video.By choosing to watch one portion,the observer misses out on the events occurring somewhere else in the spherical scene.This causes the observer to experience fear of missing out(FOMO).Hence,a novel personalized video synopsis approach for the generation of non-spherical videos has been introduced to address this issue.It also includes an action recognition module that makes it easy to display necessary actions by prioritizing them.This work minimizes and maximizes multiple goals such as loss of activity,collision,temporal consistency,length,show,and important action cost respectively.The performance of the proposed framework is evaluated through extensive simulation and compared with the state-of-art video synopsis optimization algorithms.Experimental results suggest that some constraints are better optimized by using the latest metaheuristic optimization algorithms to generate compact personalized synopsis videos from spherical surveillance videos.展开更多
Internal learning-based video inpainting methods have shown promising results by exploiting the intrinsic properties of the video to fill in the missing region without external dataset supervision.However,existing int...Internal learning-based video inpainting methods have shown promising results by exploiting the intrinsic properties of the video to fill in the missing region without external dataset supervision.However,existing internal learning-based video inpainting methods would produce inconsistent structures or blurry textures due to the insufficient utilisation of motion priors within the video sequence.In this paper,the authors propose a new internal learning-based video inpainting model called appearance consistency and motion coherence network(ACMC-Net),which can not only learn the recurrence of appearance prior but can also capture motion coherence prior to improve the quality of the inpainting results.In ACMC-Net,a transformer-based appearance network is developed to capture global context information within the video frame for representing appearance consistency accurately.Additionally,a novel motion coherence learning scheme is proposed to learn the motion prior in a video sequence effectively.Finally,the learnt internal appearance consistency and motion coherence are implicitly propagated to the missing regions to achieve inpainting well.Extensive experiments conducted on the DAVIS dataset show that the proposed model obtains the superior performance in terms of quantitative measurements and produces more visually plausible results compared with the state-of-the-art methods.展开更多
The rapid development of short video platforms poses new challenges for traditional recommendation systems.Recommender systems typically depend on two types of user behavior feedback to construct user interest profile...The rapid development of short video platforms poses new challenges for traditional recommendation systems.Recommender systems typically depend on two types of user behavior feedback to construct user interest profiles:explicit feedback(interactive behavior),which significantly influences users’short-term interests,and implicit feedback(viewing time),which substantially affects their long-term interests.However,the previous model fails to distinguish between these two feedback methods,leading it to predict only the overall preferences of users based on extensive historical behavior sequences.Consequently,it cannot differentiate between users’long-term and shortterm interests,resulting in low accuracy in describing users’interest states and predicting the evolution of their interests.This paper introduces a video recommendationmodel calledCAT-MFRec(CrossAttention Transformer-Mixed Feedback Recommendation)designed to differentiate between explicit and implicit user feedback within the DIEN(Deep Interest Evolution Network)framework.This study emphasizes the separate learning of the two types of behavioral feedback,effectively integrating them through the cross-attention mechanism.Additionally,it leverages the long sequence dependence capabilities of Transformer technology to accurately construct user interest profiles and predict the evolution of user interests.Experimental results indicate that CAT-MF Rec significantly outperforms existing recommendation methods across various performance indicators.This advancement offers new theoretical and practical insights for the development of video recommendations,particularly in addressing complex and dynamic user behavior patterns.展开更多
Airway management plays a crucial role in providing adequate oxygenation and ventilation to patients during various medical procedures and emergencies.When patients have a limited mouth opening due to factors such as ...Airway management plays a crucial role in providing adequate oxygenation and ventilation to patients during various medical procedures and emergencies.When patients have a limited mouth opening due to factors such as trauma,inflammation,or anatomical abnormalities airway management becomes challenging.A commonly utilized method to overcome this challenge is the use of video laryngoscopy(VL),which employs a specialized device equipped with a camera and a light source to allow a clear view of the larynx and vocal cords.VL overcomes the limitations of direct laryngoscopy in patients with limited mouth opening,enabling better visualization and successful intubation.Various types of VL blades are available.We devised a novel flangeless video laryngoscope for use in patients with a limited mouth opening and then tested it on a manikin.展开更多
Semantic segmentation is a core task in computer vision that allows AI models to interact and understand their surrounding environment. Similarly to how humans subconsciously segment scenes, this ability is crucial fo...Semantic segmentation is a core task in computer vision that allows AI models to interact and understand their surrounding environment. Similarly to how humans subconsciously segment scenes, this ability is crucial for scene understanding. However, a challenge many semantic learning models face is the lack of data. Existing video datasets are limited to short, low-resolution videos that are not representative of real-world examples. Thus, one of our key contributions is a customized semantic segmentation version of the Walking Tours Dataset that features hour-long, high-resolution, real-world data from tours of different cities. Additionally, we evaluate the performance of open-vocabulary, semantic model OpenSeeD on our own custom dataset and discuss future implications.展开更多
Objective: The purpose of this study was to evaluate health education using videos and leaflets for preconception care (PCC) awareness among adolescent females up to six months after the health education. Methods: The...Objective: The purpose of this study was to evaluate health education using videos and leaflets for preconception care (PCC) awareness among adolescent females up to six months after the health education. Methods: The subjects were female university students living in the Kinki area. A longitudinal survey was conducted on 67 members in the intervention group, who received the health education, and 52 members in the control group, who did not receive the health education. The primary outcome measures were knowledge of PCC and the subscales of the Health Promotion Lifestyle Profile. Surveys were conducted before, after, and six months after the intervention in the intervention group, and an initial survey and survey six months later were conducted in the control group. Cochran’s Q test, Bonferroni’s multiple comparison test, and McNemar’s test were used to analyze the knowledge of PCC data. The Health Awareness, Nutrition, and Stress Management subscales of the Health Promotion Lifestyle Profile were analyzed by paired t-test, and comparisons between the intervention and control groups were performed using the two-way repeated measures analysis of variance. Results: In the intervention group of 67 people, the number of subjects who answered “correct” for five of the nine items concerning knowledge of PCC increased immediately after the health education (P = 0.006) but decreased for five items from immediately after the health education to six months later (P = 0.043). In addition, the number of respondents who answered “correct” for “low birth weight infants and future lifestyle-related diseases” (P = 0.016) increased after six months compared with before the health education. For the 52 subjects in the control group, there was no change in the number of subjects who answered “correct” for eight out of the nine items after six months. There was also no increase in scores for the Health Promotion Lifestyle Profile after six months for either the intervention or control group. Conclusion: Providing health education about PCC using videos and leaflets to adolescent females was shown to enhance the knowledge of PCC immediately after the education.展开更多
Objective:The objective of this study is to determine the effect of nurse-led instructional video(NLIV)on anxiety,satisfaction,and recovery among mothers admitted for cesarean section(CS).Materials and Methods:A quasi...Objective:The objective of this study is to determine the effect of nurse-led instructional video(NLIV)on anxiety,satisfaction,and recovery among mothers admitted for cesarean section(CS).Materials and Methods:A quasi-experimental design was carried out on the mothers scheduled for CS.Eighty participants were selected by a purposive sampling technique,which were divided(40 participants in each group)into an experimental group and a control group.Nurse-led informational video(NLIV)was shown to the experimental group,and routine care was provided for the control group.Modified hospital anxiety scale(HADS),scale for measuring maternal satisfaction in cesarean birth,and obstetric quality of recovery following cesarean delivery were used to assess anxiety,satisfaction,and recovery.Results:Both the experimental and control groups showed significant reductions in anxiety by the first postintervention day(P<0.001),with the experimental group experiencing a greater mean reduction(mean difference[MD]=4.37)than the control group(MD=3.35)but the intergroup difference was not statistically significant(P>0.05).The experimental group reported significantly higher satisfaction scores(175.55±9.42)on the 3rd postoperative day compared to the control group(151.93±14.89;P<0.001).Similarly,the experimental group’s recovery scores(79.90±6.24)were considerably higher than those of the control group(62.45±15.18;P<0.001).On the 3rd postintervention day,satisfaction was significantly associated with age(P<0.001),and recovery with gravidity(P<0.05).Conclusions:NLIV can be used in the preoperative period to reduce anxiety related to CS and to improve satisfaction and recovery after the CS.展开更多
Objective While there is consensus regarding a positive effect of video gaming on dexterity,little is known regarding how much traditional laparoscopic practice can or should be substituted with video gaming.This stud...Objective While there is consensus regarding a positive effect of video gaming on dexterity,little is known regarding how much traditional laparoscopic practice can or should be substituted with video gaming.This study was designed to assess the effects of varying the amount of traditional practice in a lap box trainer and video gaming on performance in two fundamentals of laparoscopic surgery core tasks.Methods Undergraduate and medical students were recruited and randomized into one of four groups:a control group,a lap box group,a video game group,and a combined group with 50%of the time allocated to each modality.Performance in the peg transfer and precision cutting tasks was assessed both prior to and following the 6 training sessions.Results Peg transfer performance significantly improved in the lap box group(168.4±70.6 s vs.332.9±178.2 s,p<0.001),video game group(176.7±53.3 s vs.300.0±101.2 s,p<0.001)and combined group(214.2±86.9 s vs.406.8±239.5 s,p=0.002)after training.Similar improvements were also observed in precision cutting performance in the lap box group(413.1±138.4 s vs.614.3±211.4 s,p=0.002),video game group(434.1±150.8 s vs.609.2±233.2 s,p=0.007)and combined group(469.2±185.3 s vs.663.8±296.3 s,p=0.020).When analyzing improvements in performance across three different training groups compared with the control group,we found that both the lap box group(p<0.001)and the combined group(p<0.001)showed better improvement in both tasks,and the video game group had significantly better outcomes in the precision cutting task(p=0.003).Conclusion Traditional lap box training remains the most effective method for improving the performance of simulated laparoscopic surgery.Video games can be encouraged to enhance skills retention and supplement simulated practice outside of a formal training curriculum.展开更多
The Double Take column looks at a single topic from an African and Chinese perspective.This month,we explore how we can cope with the influence of short videos.
Objectives:People living with myalgic encephalomyelitis/chronic fatigue syndrome(ME/CFS)may face unique barriers to physical activity.Active video games may overcome these barriers and increase physical activity.The p...Objectives:People living with myalgic encephalomyelitis/chronic fatigue syndrome(ME/CFS)may face unique barriers to physical activity.Active video games may overcome these barriers and increase physical activity.The primary aim of this pilot study was to determine the feasibility and acceptability of active video games to increase physical activity levels of people with ME/CFS.Methods:A mixed method design was employed.Adults living with ME/CFS were randomised to a six-month intervention of(1)pacing,(2)pacing and conventional physical activity,or(3)pacing and active video gaming.Feasibility and acceptability were determined through semi-structured interviews.Health-related outcomes(e.g.,physical activity,blood samples,quality of life,and functioning)were also collected.Results:Fifteen people were assigned an intervention group with 12 completing.Three themes were identified from post-intervention interviews(1)positivity led to overall acceptability,(2)flexibility enabled participant autonomy,and(3)knowledge was gained about the self.No changes were seen in physical activity levels.An active video gaming intervention is acceptable but not feasible for people living with ME/CFS.Conclusions:Taken together,results suggest that physical activity can be managed safely by some people living with ME/CFS.Trial registration ACTRN12616000285459.展开更多
In the wave of internet culture,short videos have become an indispensable medium for social communication.The metaphorical hot words contained within them serve as a unique linguistic phenomenon that leads topics and ...In the wave of internet culture,short videos have become an indispensable medium for social communication.The metaphorical hot words contained within them serve as a unique linguistic phenomenon that leads topics and focuses attention,greatly enriching the expressive layers and rhetorical charm of short videos,and significantly enhancing the video’s theme orientation and emotional identification.This research aims to explore the relationship between the use of metaphorical Internet buzzwords in short videos and the thematic and emotional orientation.The study adopts a combination of qualitative and quantitative methods,taking 10 videos with over 10,000 likes posted by a well-known blogger on Xiaohongshu in 2024 as the research object,transcribing the text,forming research corpora,and conducting multi-dimensional cognitive analysis on them.The study shows that about half of short videos contain metaphorical hot words.Different types of metaphorical hot words can trigger different emotional reactions from fans,especially humorous metaphorical hot words that can stimulate fans’emotional identification and resonance.In addition,in terms of fan participation,videos using metaphorical hot words tend to attract more fan attention than those that do not:these videos not only attract more fans to watch and like,but also trigger more comments and sharing behaviors.In summary,short videos cleverly use metaphors to create internet hot words,significantly enhancing the video’s thematic guidance and emotional resonance,manifested in creating popular topics,clarifying guiding themes,enhancing content attractiveness,and stimulating strong emotional identification,thereby promoting interactive behaviors such as likes and shares.These findings provide a reference for research in related fields such as metaphor,communication studies,and sociology.展开更多
Personal video recorders (PVRs) have altered the way users consume television (TV) content by allowing users to record programs and watch them at their convenience, overcoming the constraints of live broadcasting. How...Personal video recorders (PVRs) have altered the way users consume television (TV) content by allowing users to record programs and watch them at their convenience, overcoming the constraints of live broadcasting. However, standalone PVRs are limited by their individual storage capacities, restricting the number of programs they can store. While online catch-up TV services such as Hulu and Netflix mitigate this limitation by offering on-demand access to broadcast programs shortly after their initial broadcast, they require substantial storage and network resources, leading to significant infrastructural costs for service providers. To address these challenges, we propose a collaborative TV content recording system that leverages distributed PVRs, combining their storage into a virtual shared pool without additional costs. Our system aims to support all concurrent playback requests without service interruption while ensuring program availability comparable to that of local devices. The main contributions of our proposed system are fourfold. First, by sharing storage and upload bandwidth among PVRs, our system significantly expands the overall recording capacity and enables simultaneous recording of multiple programs without the physical constraints of standalone devices. Second, by utilizing erasure coding efficiently, our system reduces the storage space required for each program, allowing more programs to be recorded compared to traditional replication. Third, we propose an adaptive redundancy scheme to control the degree of redundancy of each program based on its evolving playback demand, ensuring high-quality playback by providing sufficient bandwidth for popular programs. Finally, we introduce a contribution-based incentive policy that encourages PVRs to actively participate by contributing resources, while discouraging excessive consumption of the combined storage pool. Through extensive experiments, we demonstrate the effectiveness of our proposed collaborative TV program recording system in terms of storage efficiency and performance.展开更多
BACKGROUND Video capsule endoscopy(VCE)is a noninvasive technique used to examine small bowel abnormalities in both adults and children.However,manual review of VCE images is time-consuming and labor-intensive,making ...BACKGROUND Video capsule endoscopy(VCE)is a noninvasive technique used to examine small bowel abnormalities in both adults and children.However,manual review of VCE images is time-consuming and labor-intensive,making it crucial to develop deep learning methods to assist in image analysis.AIM To employ deep learning models for the automatic classification of small bowel lesions using pediatric VCE images.METHODS We retrospectively analyzed VCE images from 162 pediatric patients who underwent VCE between January 2021 and December 2023 at the Children's Hospital of Nanjing Medical University.A total of 2298 high-resolution images were extracted,including normal mucosa and lesions(erosions/erythema,ulcers,and polyps).The images were split into training and test datasets in a 4:1 ratio.Four deep learning models:DenseNet121,Visual geometry group-16,ResNet50,and vision transformer were trained using 5-fold cross-validation,with hyperparameters adjusted for optimal classification performance.The models were evaluated based on accuracy,precision,recall,F1-score,and area under the receiver operating curve(AU-ROC).Lesion visualization was performed using gradient-weighted class activation mapping.RESULTS Abdominal pain was the most common indication for VCE,accounting for 62%of cases,followed by diarrhea,vomiting,and gastrointestinal bleeding.Abnormal lesions were detected in 93 children,with 38 diagnosed with inflammatory bowel disease.Among the deep learning models,DenseNet121 and ResNet50 demonstrated excellent classification performance,achieving accuracies of 90.6%[95%confidence interval(CI):89.2-92.0]and 90.5%(95%CI:89.9-91.2),respectively.The AU-ROC values for these models were 93.7%(95%CI:92.9-94.5)for DenseNet121 and 93.4%(95%CI:93.1-93.8)for ResNet50.CONCLUSION Our deep learning-based diagnostic tool developed in this study effectively classified lesions in pediatric VCE images,contributing to more accurate diagnoses and increased diagnostic efficiency.展开更多
Dance video is an innovative form that integrates dance art with imaging technology,enriching the expression of dance through techniques such as photography,videography,and special effects.This paper explores the defi...Dance video is an innovative form that integrates dance art with imaging technology,enriching the expression of dance through techniques such as photography,videography,and special effects.This paper explores the definition,current development,artistic expression,social impact,and future trends of dance video.Through narrative construction,visual impact,emotional resonance,technological innovation,and cultural expression,dance video enhances the narrative and visual appeal of dance.In the future,dance video will focus more on the integration of virtual reality(VR)and augmented reality(AR)technologies.However,it also faces challenges such as rapid technological updates,maintaining artistic originality,and balancing commercial interests.展开更多
基金supported by the National Natural Science Foundation of China (Nos. NSFC 61925105, 62322109, 62171257 and U22B2001)the Xplorer Prize in Information and Electronics technologiesthe Tsinghua University (Department of Electronic Engineering)-Nantong Research Institute for Advanced Communication Technologies Joint Research Center for Space, Air, Ground and Sea Cooperative Communication Network Technology
文摘Multimedia semantic communication has been receiving increasing attention due to its significant enhancement of communication efficiency.Semantic coding,which is oriented towards extracting and encoding the key semantics of video for transmission,is a key aspect in the framework of multimedia semantic communication.In this paper,we propose a facial video semantic coding method with low bitrate based on the temporal continuity of video semantics.At the sender’s end,we selectively transmit facial keypoints and deformation information,allocating distinct bitrates to different keypoints across frames.Compressive techniques involving sampling and quantization are employed to reduce the bitrate while retaining facial key semantic information.At the receiver’s end,a GAN-based generative network is utilized for reconstruction,effectively mitigating block artifacts and buffering problems present in traditional codec algorithms under low bitrates.The performance of the proposed approach is validated on multiple datasets,such as VoxCeleb and TalkingHead-1kH,employing metrics such as LPIPS,DISTS,and AKD for assessment.Experimental results demonstrate significant advantages over traditional codec methods,achieving up to approximately 10-fold bitrate reduction in prolonged,stable head pose scenarios across diverse conversational video settings.
文摘The application of short videos in agricultural scenarios has become a new form of productive force driving agricultural development,injecting new vitality and opportunities into traditional agriculture.These videos leverage the unique expressive logic of the platform by adopting a small entry point and prioritizing dissemination rate.They are strategically planned in terms of content,visuals,and interaction to cater to users needs for relaxation,knowledge acquisition,social sharing,agricultural product marketing,and talent display.Through careful design,full creativity,rich emotion,and the creation of distinct character personalities,these videos deliver positive,entertaining,informative,and opinion-driven agricultural content.The production and operation of agricultural short videos can be effectively optimized by analyzing the characteristics of both popular and less popular videos,and utilizing smart tools and trending topics.
文摘Objectives:Medical students often rely on recreational internet media to relieve the stress caused by immense academic and life pressures,and among these media,short-form videos,which are an emerging digital medium,have gradually become the mainstream choice of students to relieve their stress.However,the addiction caused by their usage has attracted the widespread attention of both academia and society,which is why the purpose of this study is to systematically explore the underlying mechanisms that link perceived stress,entertainment gratification,emotional gratification,short-form video usage intensity,and short-form video addiction based on multiple theoretical frameworks including the Compensatory Internet Use Model(CIU),the Interaction of Person-Affect-Cognition-Execution Model(I-PACE),and the Use and Gratification Theory(UGT).Methods:A hypothetical model with 9 research hypotheses was constructed.Taking medical students from Chi-nese universities as the research subjects,1057 valid responses were collected through an online questionnaire survey,including 358 males and 658 females.Structural equation modelling(SEM)was performed using the AMOS software to test the research hypotheses.Results:(1)Perceived stress positively predicted entertainment gratification and emotional gratification(β=0.72,p<0.001;β=0.61,p<0.001);(2)Entertainment gratifi-cation and emotional gratification positively influenced short-form video usage intensity(β=0.35,p<0.001;β=0.19,p<0.001);(3)Entertainment gratification and emotional gratification positively predicted short-form video addiction(β=0.40,p<0.001;β=0.17,p<0.001);(4)Short-form video usage intensity positively influenced short-form video addiction(β=0.36,p<0.001);and(5)Perceived stress exerted an indirect but positive effect on both short-form video usage intensity and short-form video addiction,mediated by entertainment and emotional gratification(β=0.37,p<0.001;β=0.52,p<0.001).Conclusion:The mechanisms that underlie medical students’short-form video addiction in stressful situations were revealed in this study.It was found that stress enhances medical students’need for entertainment and emotional online compensation,prompting more frequent short-form video usage and ultimately leading to addiction.These results underscore the need to address the stressors faced by medical students.Effective interventions should prioritise stress management strategies and promote healthier alternative coping mechanisms to mitigate the risk of addiction.
基金supported by the Zhejiang Provincial Natural Science Foundation of China(No.LQ23F030001)the National Natural Science Foundation of China(No.62406280)+5 种基金the Autism Research Special Fund of Zhejiang Foundation for Disabled Persons(No.2023008)the Liaoning Province Higher Education Innovative Talents Program Support Project(No.LR2019058)the Liaoning Province Joint Open Fund for Key Scientific and Technological Innovation Bases(No.2021-KF-12-05)the Central Guidance on Local Science and Technology Development Fund of Liaoning Province(No.2023JH6/100100066)the Key Laboratory for Biomedical Engineering of Ministry of Education,Zhejiang University,Chinain part by the Open Research Fund of the State Key Laboratory of Cognitive Neuroscience and Learning.
文摘Video action recognition(VAR)aims to analyze dynamic behaviors in videos and achieve semantic understanding.VAR faces challenges such as temporal dynamics,action-scene coupling,and the complexity of human interactions.Existing methods can be categorized into motion-level,event-level,and story-level ones based on spatiotemporal granularity.However,single-modal approaches struggle to capture complex behavioral semantics and human factors.Therefore,in recent years,vision-language models(VLMs)have been introduced into this field,providing new research perspectives for VAR.In this paper,we systematically review spatiotemporal hierarchical methods in VAR and explore how the introduction of large models has advanced the field.Additionally,we propose the concept of“Factor”to identify and integrate key information from both visual and textual modalities,enhancing multimodal alignment.We also summarize various multimodal alignment methods and provide in-depth analysis and insights into future research directions.
基金supported by the International Joint Research Project of Huiyan International College,Faculty of Education,Beijing Normal University(Grant Number:ICER202102).
文摘Objectives:Short video addiction has emerged as a significant public health issue in recent years,with a growing trend toward severity.However,research on the causes and impacts of short video addiction remains limited,and understanding of the variable“TikTok brain”is still in its infancy.Therefore,based on the Stimulus-Organism-Behavior-Consequence(SOBC)framework,we proposed six research hypotheses and constructed a model to explore the relationships between short video usage intensity,TikTok brain,short video addiction,and decreased attention control.Methods:Given that students are considered a high-risk group for excessive short video use,we collected 1086 valid participants from Chinese student users,including 609 males(56.1%)and 477 females(43.9%),with an average participant age of 19.84 years,to test the hypotheses.Results:(1)Short video usage intensity was positively related to short video addiction,TikTok brain,and decreased attention control;(2)TikTok brain was positively related to short video addiction and decreased attention control;and(3)Short video addiction was positively related to decreased attention control.Conclusions:These findings suggest that although excessive use of short video applications brings negative consequences,users still spend significant amounts of time on these platforms,indicating a need for strict self-regulation of usage time.
文摘Video synopsis is an effective way to easily summarize long-recorded surveillance videos.The omnidirectional view allows the observer to select the desired fields of view(FoV)from the different FoVavailable for spherical surveillance video.By choosing to watch one portion,the observer misses out on the events occurring somewhere else in the spherical scene.This causes the observer to experience fear of missing out(FOMO).Hence,a novel personalized video synopsis approach for the generation of non-spherical videos has been introduced to address this issue.It also includes an action recognition module that makes it easy to display necessary actions by prioritizing them.This work minimizes and maximizes multiple goals such as loss of activity,collision,temporal consistency,length,show,and important action cost respectively.The performance of the proposed framework is evaluated through extensive simulation and compared with the state-of-art video synopsis optimization algorithms.Experimental results suggest that some constraints are better optimized by using the latest metaheuristic optimization algorithms to generate compact personalized synopsis videos from spherical surveillance videos.
基金Shenzhen Science and Technology Programme,Grant/Award Number:JCYJ202308071208000012023 Shenzhen sustainable supporting funds for colleges and universities,Grant/Award Number:20231121165240001Guangdong Provincial Key Laboratory of Ultra High Definition Immersive Media Technology,Grant/Award Number:2024B1212010006。
文摘Internal learning-based video inpainting methods have shown promising results by exploiting the intrinsic properties of the video to fill in the missing region without external dataset supervision.However,existing internal learning-based video inpainting methods would produce inconsistent structures or blurry textures due to the insufficient utilisation of motion priors within the video sequence.In this paper,the authors propose a new internal learning-based video inpainting model called appearance consistency and motion coherence network(ACMC-Net),which can not only learn the recurrence of appearance prior but can also capture motion coherence prior to improve the quality of the inpainting results.In ACMC-Net,a transformer-based appearance network is developed to capture global context information within the video frame for representing appearance consistency accurately.Additionally,a novel motion coherence learning scheme is proposed to learn the motion prior in a video sequence effectively.Finally,the learnt internal appearance consistency and motion coherence are implicitly propagated to the missing regions to achieve inpainting well.Extensive experiments conducted on the DAVIS dataset show that the proposed model obtains the superior performance in terms of quantitative measurements and produces more visually plausible results compared with the state-of-the-art methods.
基金supported by National Natural Science Foundation of China(62072416)Key Research and Development Special Project of Henan Province(221111210500)Key TechnologiesR&DProgram of Henan rovince(232102211053,242102211071).
文摘The rapid development of short video platforms poses new challenges for traditional recommendation systems.Recommender systems typically depend on two types of user behavior feedback to construct user interest profiles:explicit feedback(interactive behavior),which significantly influences users’short-term interests,and implicit feedback(viewing time),which substantially affects their long-term interests.However,the previous model fails to distinguish between these two feedback methods,leading it to predict only the overall preferences of users based on extensive historical behavior sequences.Consequently,it cannot differentiate between users’long-term and shortterm interests,resulting in low accuracy in describing users’interest states and predicting the evolution of their interests.This paper introduces a video recommendationmodel calledCAT-MFRec(CrossAttention Transformer-Mixed Feedback Recommendation)designed to differentiate between explicit and implicit user feedback within the DIEN(Deep Interest Evolution Network)framework.This study emphasizes the separate learning of the two types of behavioral feedback,effectively integrating them through the cross-attention mechanism.Additionally,it leverages the long sequence dependence capabilities of Transformer technology to accurately construct user interest profiles and predict the evolution of user interests.Experimental results indicate that CAT-MF Rec significantly outperforms existing recommendation methods across various performance indicators.This advancement offers new theoretical and practical insights for the development of video recommendations,particularly in addressing complex and dynamic user behavior patterns.
文摘Airway management plays a crucial role in providing adequate oxygenation and ventilation to patients during various medical procedures and emergencies.When patients have a limited mouth opening due to factors such as trauma,inflammation,or anatomical abnormalities airway management becomes challenging.A commonly utilized method to overcome this challenge is the use of video laryngoscopy(VL),which employs a specialized device equipped with a camera and a light source to allow a clear view of the larynx and vocal cords.VL overcomes the limitations of direct laryngoscopy in patients with limited mouth opening,enabling better visualization and successful intubation.Various types of VL blades are available.We devised a novel flangeless video laryngoscope for use in patients with a limited mouth opening and then tested it on a manikin.
文摘Semantic segmentation is a core task in computer vision that allows AI models to interact and understand their surrounding environment. Similarly to how humans subconsciously segment scenes, this ability is crucial for scene understanding. However, a challenge many semantic learning models face is the lack of data. Existing video datasets are limited to short, low-resolution videos that are not representative of real-world examples. Thus, one of our key contributions is a customized semantic segmentation version of the Walking Tours Dataset that features hour-long, high-resolution, real-world data from tours of different cities. Additionally, we evaluate the performance of open-vocabulary, semantic model OpenSeeD on our own custom dataset and discuss future implications.
文摘Objective: The purpose of this study was to evaluate health education using videos and leaflets for preconception care (PCC) awareness among adolescent females up to six months after the health education. Methods: The subjects were female university students living in the Kinki area. A longitudinal survey was conducted on 67 members in the intervention group, who received the health education, and 52 members in the control group, who did not receive the health education. The primary outcome measures were knowledge of PCC and the subscales of the Health Promotion Lifestyle Profile. Surveys were conducted before, after, and six months after the intervention in the intervention group, and an initial survey and survey six months later were conducted in the control group. Cochran’s Q test, Bonferroni’s multiple comparison test, and McNemar’s test were used to analyze the knowledge of PCC data. The Health Awareness, Nutrition, and Stress Management subscales of the Health Promotion Lifestyle Profile were analyzed by paired t-test, and comparisons between the intervention and control groups were performed using the two-way repeated measures analysis of variance. Results: In the intervention group of 67 people, the number of subjects who answered “correct” for five of the nine items concerning knowledge of PCC increased immediately after the health education (P = 0.006) but decreased for five items from immediately after the health education to six months later (P = 0.043). In addition, the number of respondents who answered “correct” for “low birth weight infants and future lifestyle-related diseases” (P = 0.016) increased after six months compared with before the health education. For the 52 subjects in the control group, there was no change in the number of subjects who answered “correct” for eight out of the nine items after six months. There was also no increase in scores for the Health Promotion Lifestyle Profile after six months for either the intervention or control group. Conclusion: Providing health education about PCC using videos and leaflets to adolescent females was shown to enhance the knowledge of PCC immediately after the education.
文摘Objective:The objective of this study is to determine the effect of nurse-led instructional video(NLIV)on anxiety,satisfaction,and recovery among mothers admitted for cesarean section(CS).Materials and Methods:A quasi-experimental design was carried out on the mothers scheduled for CS.Eighty participants were selected by a purposive sampling technique,which were divided(40 participants in each group)into an experimental group and a control group.Nurse-led informational video(NLIV)was shown to the experimental group,and routine care was provided for the control group.Modified hospital anxiety scale(HADS),scale for measuring maternal satisfaction in cesarean birth,and obstetric quality of recovery following cesarean delivery were used to assess anxiety,satisfaction,and recovery.Results:Both the experimental and control groups showed significant reductions in anxiety by the first postintervention day(P<0.001),with the experimental group experiencing a greater mean reduction(mean difference[MD]=4.37)than the control group(MD=3.35)but the intergroup difference was not statistically significant(P>0.05).The experimental group reported significantly higher satisfaction scores(175.55±9.42)on the 3rd postoperative day compared to the control group(151.93±14.89;P<0.001).Similarly,the experimental group’s recovery scores(79.90±6.24)were considerably higher than those of the control group(62.45±15.18;P<0.001).On the 3rd postintervention day,satisfaction was significantly associated with age(P<0.001),and recovery with gravidity(P<0.05).Conclusions:NLIV can be used in the preoperative period to reduce anxiety related to CS and to improve satisfaction and recovery after the CS.
基金the financial support from the China Scholarship Council(Grant No.202106370009)and Alberta Innovate Graduate Student Scholarship.
文摘Objective While there is consensus regarding a positive effect of video gaming on dexterity,little is known regarding how much traditional laparoscopic practice can or should be substituted with video gaming.This study was designed to assess the effects of varying the amount of traditional practice in a lap box trainer and video gaming on performance in two fundamentals of laparoscopic surgery core tasks.Methods Undergraduate and medical students were recruited and randomized into one of four groups:a control group,a lap box group,a video game group,and a combined group with 50%of the time allocated to each modality.Performance in the peg transfer and precision cutting tasks was assessed both prior to and following the 6 training sessions.Results Peg transfer performance significantly improved in the lap box group(168.4±70.6 s vs.332.9±178.2 s,p<0.001),video game group(176.7±53.3 s vs.300.0±101.2 s,p<0.001)and combined group(214.2±86.9 s vs.406.8±239.5 s,p=0.002)after training.Similar improvements were also observed in precision cutting performance in the lap box group(413.1±138.4 s vs.614.3±211.4 s,p=0.002),video game group(434.1±150.8 s vs.609.2±233.2 s,p=0.007)and combined group(469.2±185.3 s vs.663.8±296.3 s,p=0.020).When analyzing improvements in performance across three different training groups compared with the control group,we found that both the lap box group(p<0.001)and the combined group(p<0.001)showed better improvement in both tasks,and the video game group had significantly better outcomes in the precision cutting task(p=0.003).Conclusion Traditional lap box training remains the most effective method for improving the performance of simulated laparoscopic surgery.Video games can be encouraged to enhance skills retention and supplement simulated practice outside of a formal training curriculum.
文摘The Double Take column looks at a single topic from an African and Chinese perspective.This month,we explore how we can cope with the influence of short videos.
基金funded by the Mason Foundation National Medical Program(MAS2015F053).
文摘Objectives:People living with myalgic encephalomyelitis/chronic fatigue syndrome(ME/CFS)may face unique barriers to physical activity.Active video games may overcome these barriers and increase physical activity.The primary aim of this pilot study was to determine the feasibility and acceptability of active video games to increase physical activity levels of people with ME/CFS.Methods:A mixed method design was employed.Adults living with ME/CFS were randomised to a six-month intervention of(1)pacing,(2)pacing and conventional physical activity,or(3)pacing and active video gaming.Feasibility and acceptability were determined through semi-structured interviews.Health-related outcomes(e.g.,physical activity,blood samples,quality of life,and functioning)were also collected.Results:Fifteen people were assigned an intervention group with 12 completing.Three themes were identified from post-intervention interviews(1)positivity led to overall acceptability,(2)flexibility enabled participant autonomy,and(3)knowledge was gained about the self.No changes were seen in physical activity levels.An active video gaming intervention is acceptable but not feasible for people living with ME/CFS.Conclusions:Taken together,results suggest that physical activity can be managed safely by some people living with ME/CFS.Trial registration ACTRN12616000285459.
文摘In the wave of internet culture,short videos have become an indispensable medium for social communication.The metaphorical hot words contained within them serve as a unique linguistic phenomenon that leads topics and focuses attention,greatly enriching the expressive layers and rhetorical charm of short videos,and significantly enhancing the video’s theme orientation and emotional identification.This research aims to explore the relationship between the use of metaphorical Internet buzzwords in short videos and the thematic and emotional orientation.The study adopts a combination of qualitative and quantitative methods,taking 10 videos with over 10,000 likes posted by a well-known blogger on Xiaohongshu in 2024 as the research object,transcribing the text,forming research corpora,and conducting multi-dimensional cognitive analysis on them.The study shows that about half of short videos contain metaphorical hot words.Different types of metaphorical hot words can trigger different emotional reactions from fans,especially humorous metaphorical hot words that can stimulate fans’emotional identification and resonance.In addition,in terms of fan participation,videos using metaphorical hot words tend to attract more fan attention than those that do not:these videos not only attract more fans to watch and like,but also trigger more comments and sharing behaviors.In summary,short videos cleverly use metaphors to create internet hot words,significantly enhancing the video’s thematic guidance and emotional resonance,manifested in creating popular topics,clarifying guiding themes,enhancing content attractiveness,and stimulating strong emotional identification,thereby promoting interactive behaviors such as likes and shares.These findings provide a reference for research in related fields such as metaphor,communication studies,and sociology.
基金supported by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(Nos.2019R1A2C1002221 and RS-2023-00252186)Institute of Information&Communications Technology Planning&Evaluation(IITP)grant funded by the Korea government(MSIT)(Nos.2021-0-00590,RS-2021-II210590Decentralized High Performance Consensus for Large-Scale Blockchains).
文摘Personal video recorders (PVRs) have altered the way users consume television (TV) content by allowing users to record programs and watch them at their convenience, overcoming the constraints of live broadcasting. However, standalone PVRs are limited by their individual storage capacities, restricting the number of programs they can store. While online catch-up TV services such as Hulu and Netflix mitigate this limitation by offering on-demand access to broadcast programs shortly after their initial broadcast, they require substantial storage and network resources, leading to significant infrastructural costs for service providers. To address these challenges, we propose a collaborative TV content recording system that leverages distributed PVRs, combining their storage into a virtual shared pool without additional costs. Our system aims to support all concurrent playback requests without service interruption while ensuring program availability comparable to that of local devices. The main contributions of our proposed system are fourfold. First, by sharing storage and upload bandwidth among PVRs, our system significantly expands the overall recording capacity and enables simultaneous recording of multiple programs without the physical constraints of standalone devices. Second, by utilizing erasure coding efficiently, our system reduces the storage space required for each program, allowing more programs to be recorded compared to traditional replication. Third, we propose an adaptive redundancy scheme to control the degree of redundancy of each program based on its evolving playback demand, ensuring high-quality playback by providing sufficient bandwidth for popular programs. Finally, we introduce a contribution-based incentive policy that encourages PVRs to actively participate by contributing resources, while discouraging excessive consumption of the combined storage pool. Through extensive experiments, we demonstrate the effectiveness of our proposed collaborative TV program recording system in terms of storage efficiency and performance.
文摘BACKGROUND Video capsule endoscopy(VCE)is a noninvasive technique used to examine small bowel abnormalities in both adults and children.However,manual review of VCE images is time-consuming and labor-intensive,making it crucial to develop deep learning methods to assist in image analysis.AIM To employ deep learning models for the automatic classification of small bowel lesions using pediatric VCE images.METHODS We retrospectively analyzed VCE images from 162 pediatric patients who underwent VCE between January 2021 and December 2023 at the Children's Hospital of Nanjing Medical University.A total of 2298 high-resolution images were extracted,including normal mucosa and lesions(erosions/erythema,ulcers,and polyps).The images were split into training and test datasets in a 4:1 ratio.Four deep learning models:DenseNet121,Visual geometry group-16,ResNet50,and vision transformer were trained using 5-fold cross-validation,with hyperparameters adjusted for optimal classification performance.The models were evaluated based on accuracy,precision,recall,F1-score,and area under the receiver operating curve(AU-ROC).Lesion visualization was performed using gradient-weighted class activation mapping.RESULTS Abdominal pain was the most common indication for VCE,accounting for 62%of cases,followed by diarrhea,vomiting,and gastrointestinal bleeding.Abnormal lesions were detected in 93 children,with 38 diagnosed with inflammatory bowel disease.Among the deep learning models,DenseNet121 and ResNet50 demonstrated excellent classification performance,achieving accuracies of 90.6%[95%confidence interval(CI):89.2-92.0]and 90.5%(95%CI:89.9-91.2),respectively.The AU-ROC values for these models were 93.7%(95%CI:92.9-94.5)for DenseNet121 and 93.4%(95%CI:93.1-93.8)for ResNet50.CONCLUSION Our deep learning-based diagnostic tool developed in this study effectively classified lesions in pediatric VCE images,contributing to more accurate diagnoses and increased diagnostic efficiency.
基金National Social Science Fund of the Chongqing,China(Grant No.2023PY86)。
文摘Dance video is an innovative form that integrates dance art with imaging technology,enriching the expression of dance through techniques such as photography,videography,and special effects.This paper explores the definition,current development,artistic expression,social impact,and future trends of dance video.Through narrative construction,visual impact,emotional resonance,technological innovation,and cultural expression,dance video enhances the narrative and visual appeal of dance.In the future,dance video will focus more on the integration of virtual reality(VR)and augmented reality(AR)technologies.However,it also faces challenges such as rapid technological updates,maintaining artistic originality,and balancing commercial interests.
