Search Results (534)

Short videos have become a dominant format in digital commerce, enabling brands to engage consumers and drive purchases through dynamic and visually rich content. This highlights the need for a more nuanced understanding of visual framing strategies, that is, what elements are shown and how they are presented. Drawing on Cognitive Load Theory, this study explores the impact of visual compositional framing strategies and their dynamics on consumer engagement and sales. Applying a CNN-based deep learning model, 249,043 images (video frames) extracted from 3426 book-related short sales videos on Douyin are classified into one of three categories: functional, contextual, or social, according to the visual composition of the frame. Further econometric modeling reveals distinct effects of such framing categories: functional framing is positively associated with both engagement and sales, contextual framing relates to higher sales only, while social framing relates positively to engagement but negatively to sales. From a dynamic perspective, frequent transitions between framing types within a short video increase visual complexity, which reduces both engagement and sales and moderates the effects of specific framing strategies. These findings advance theoretical understanding of visual framing in dynamic media environments and offer practical insights for designing more effective short video content. Full article

Background/Objectives: Accurate estimation of nutritional content from food images has important applications in dietary assessment and public health surveillance. While large language models (LLMs) have shown promise for this task, the effects of prompt design and model selection on estimation accuracy remain poorly characterized. Methods: We evaluated three Claude models (Haiku 4.5, Sonnet 4.6, Opus 4.6) for visual estimation of five mandatory nutritional components (energy, protein, fat, carbohydrate, and salt equivalent) across three datasets: NutriImage (691 Japanese meal photographs with dietitian-validated ground truth, after OCR-mask quality filtering), SNAPMe (1463 US meal photographs from a publicly available benchmark), and the Japan Branded Food Database (JBFD; 989–1000 packaged food product images). We systematically compared a default prompt and a visual estimation prompt explicitly instructing the model not to read any text or numbers visible in the image. Results: The visual estimation prompt substantially improved accuracy when paired with a sufficiently capable model (energy R²: 0.23 for Haiku to 0.60 for Sonnet, JBFD). Sonnet and Opus substantially outperformed Haiku across all datasets, while differences between Sonnet and Opus were small (MedAPE difference 1–3 percentage points). Packaged food images (JBFD) yielded higher R² than meal photographs. Salt equivalent showed consistently poor accuracy (MedAPE 34–64%). On SNAPMe, Sonnet achieved lower energy MAE (116.9 vs. 123.0 kcal, −4.9%) and lower MAE for protein (5.9 vs. 7.9 g, −25.7%) and fat (6.6 vs. 8.7 g, −24.5%) compared with a recent ChatGPT-5 study. Conclusions: Claude Sonnet offers the best cost-performance balance for LLM-based nutritional estimation. Prompt design substantially affects accuracy, but only when paired with a sufficiently capable model; model visual recognition capability appears to be a key determinant of performance. These findings highlight the inherent difficulty of this task and provide practical guidance for dietary assessment system development. Full article

This study aims to explore the role of digital public relations in enhancing the corporate image of educational institutions by focusing on Bayan College in the Sultanate of Oman. The study is based on a central question regarding the effectiveness of social media platforms in improving the institution’s image among its audience, particularly students. To achieve its objectives, the study employed the descriptive and analytical method using a questionnaire tool, with a sample of 662 students from various academic disciplines at the college. The results showed that Instagram was the most widely used social media platform and that digital public relations played an effective role in strengthening the college’s image. The findings also indicated no statistically significant differences attributable to gender or academic specialization, while differences were found based on academic year. The study recommends adopting effective digital communication strategies and enhancing the use of social platforms to build a positive and sustainable institutional image. Full article

The conducted study examines the environmental impacts of logistics sector operations and possible solutions to reduce them. The main aim is to balance the stability of sustainability principles across economic, social, and environmental indicators by applying green methods. Logistics warehouse operations are also important, because they can also have negative impacts, but this study focuses on environmental pressures. Logistics firms choose to implement green sustainable methods because their major sustainability aim is the protection of the environment. Moreover, by achieving this vision, logistics companies can create better brand image and attract more customers and suppliers. This study included a survey conducted among various professionals to obtain a deep understanding of the topic, with the findings being visualised in charts to improve understanding and generate an interest for this area of study; a table illustration is also provided to clearly present the factors contributing to the environmental footprint of logistics firms and solutions to mitigate them. According to the results given in this article, it can be stated that the modern world shows great interest in the topic of sustainability and takes into strict consideration green methods in order to achieve sustainable operations efficiently. Full article

Despite growing interest in destination branding, empirical evidence is still limited in explaining how brand personality influences tourist behavior mediated by integrated aspects. This study addresses this research gap by proposing and testing a structural model that considers the mediating role of tourist experience, authenticity, and trust in the relationship between brand personality and tourist behavior. The methodology used was quantitative and causal–correlational, using structural equation modeling and a sample of 514 Peruvian tourists selected through non-probabilistic convenience sampling. The results show that brand personality significantly influences tourist experience and destination authenticity, while its direct effect on trust is weak. In addition, experience positively influences trust and authenticity. Significantly, authenticity and experience show direct and positive effects on tourist behavior, while trust has a negative effect. These findings contribute to the advancement of the literature by integrating a single explanatory model with experiential, cognitive, and relational variables, broadening the understanding of the indirect role of brand personality in the tourism context. From a practical standpoint, the results suggest that destination managers should focus on enhancing brand personality and authenticity. Full article

Graphene oxide (GO) can improve cementitious materials, but its effectiveness often differs among commercial products even at the same nominal dosage. This study proposes an optical microscopy-based framework for evaluating the initial dispersion quality of commercial GO suspensions before cement mixing. Under fixed slide preparation, imaging, and image-processing conditions, optical microscopy was used as a geometric dispersion-evaluation tool rather than a direct chemical characterization method. Three image-derived features, namely projected boundary richness, coarse-agglomerate fraction, and spatial dispersion uniformity, were integrated into an optical initial dispersion quality index, $D_j$. The framework was applied to five commercial GO products at a fixed dosage of 0.03 wt% of binder. The $D_j$-based ranking was Brand 1 > Brand 2 > Brand 3 > Brand 4 > Brand 5, and remained unchanged when the coarse-agglomerate threshold varied from 20 to 100 μm². Bootstrap resampling confirmed the robustness of the ranking. The 3-day compressive strength increased from 51.1 MPa for the control mixture to 52.6~60.8 MPa for GO-modified mortars, corresponding to enhancement ratios of 3.1~19.2%. The strength-enhancement ranking was identical to the optical dispersion ranking, with a Spearman rank correlation coefficient of ${\rho }_s=1.0$. The proposed $D_j$ index provides a practical pre-mixing screening tool for comparing commercial GO products before strength testing or detailed physicochemical characterization. Full article

Digital technologies are reshaping how tourists and hospitality consumers search for, personalize, interpret, and share experiences. This study examines customer value co-creation (VCC) as a mechanism linking digital-age participation with consumer outcomes in tourism and hospitality. A PRISMA 2020-guided meta-analysis was conducted using Scopus, Web of Science Core Collection, and Hospitality & Tourism Complete. Forty peer-reviewed studies met the eligibility criteria. Random-effects models synthesized unadjusted correlations between VCC and its main antecedents and outcomes. VCC was positively associated with customer engagement, perceived innovation, and sustainability/CSR-related perceptions. On the outcome side, the strongest and most mature associations were observed for satisfaction (r = 0.64), loyalty (r = 0.61), and perceived value (r = 0.52). Extended outcomes, including experience evaluations, well-being, image, and equity-related indicators, were also positive on average but less empirically mature. High heterogeneity and wide prediction intervals show that VCC is better understood as a context-dependent mechanism rather than a universally strong predictor. Exploratory evidence suggests that digitally intensive service environments may strengthen the VCC–loyalty association. Although the evidence base is not cultural-tourism-specific, the findings are relevant to cultural and heritage settings where digital touchpoints can support interpretation, perceived authenticity, symbolic meaning, and post-visit advocacy. Full article

Aspect-Based Sentiment Analysis explores sentiment trends related to specific opinion aspects and holds significant commercial potential for monitoring brand reputation, understanding customer satisfaction, and personalizing recommendations. However, traditional methods rely exclusively on textual input and often struggle when the target aspect is not mentioned in the sentence. Multimodal Aspect-Based Sentiment Analysis addresses this limitation by incorporating both textual and visual modalities to enable more comprehensive sentiment understanding. Despite advancements in deep learning and transformer-based architectures, existing models often suffer from suboptimal modality fusion and weak aspect grounding, limiting their classification accuracy. To overcome these challenges, we propose an Attention-Driven Feature Fusion (ADFF) approach based on a three-stage hierarchical attention mechanism. First, it only fuses text and image embeddings. Second, it incorporates aspect-level features. Third, a multi-head attention layer further enhances cross-modal dependencies. The resulting representation is passed to a Long Short-Term Memory (LSTM) classifier for sentiment polarity prediction. We evaluate our model on three benchmark datasets, namely Twitter-2015, Twitter-2017, and MASAD. The experimental results demonstrate that the proposed model substantially outperforms state-of-the-art multimodal and unimodal baselines, improves both accuracy and F1-score, achieving 82.55% accuracy and 81.05% F1-score on Twitter-2015, 77.07% accuracy and 77.15% F1-score on Twitter-2017, and up to 99.67% accuracy and F1-score in the Plant domain of MASAD, where we observe consistent improvements across all seven domains. These results highlight the effectiveness and scalability of the hierarchical attention-based fusion strategy for real-world aspect-based sentiment analysis tasks. Full article

This paper presents a deterministic system that combines textual semantic data from panoramic images with LiDAR point clouds in a mobile mapping setup. Urban scenes often include textual elements, such as signs and business names, that provide key details typically missing from LiDAR-based urban digital twins. The presented method uses deep learning-based OCR to extract text from street panoramas and then categorizes it into urban types using a rule-based classifier. Text regions are geometrically projected into the LiDAR environment by converting image coordinates into viewing rays that intersect LiDAR surfaces, such as facades. Data from multiple panoramas are merged with confidence-weighted spatial clustering to produce consistent semantic markers for urban features. Extracted business names enable text-based searches of the LiDAR point cloud, allowing facility location by category, keyword, or brand. Tests on datasets from European and U.S. cities support plausible facade-level localization and demonstrate the framework’s ability to enhance LiDAR point clouds with searchable semantic information. The main contribution is not a new standalone OCR or LiDAR-processing algorithm, but a deterministic multimodal integration framework that combines deep-learning OCR, geometric back-projection, and cross-view spatial fusion to convert street-level textual cues into reliable, queryable 3D semantic markers within mobile-mapping LiDAR data. Full article

With the development of the Internet of Things (IoT) and microelectronics technology, the methods used to tamper with fuel dispensers have become increasingly concealed, posing significant challenges to market supervision and law enforcement. This paper releases a tampering features dataset of assembled printed circuit boards (PCBs) from fuel dispensers, aiming to provide high-quality data support for automated, computer-vision-based illicit metering detection. The dataset encompasses multi-class tampering features derived from 189 high-resolution images of PCBs seized during real-world law enforcement, covering 5 mainstream brands. To eliminate perspective bias, rigorous lens distortion correction and four-point homography transformation preprocessing were conducted on the images. Additionally, six typical tampering features (e.g., the addition of tampered surface-mount resistors) were manually and precisely annotated, and then cross-checked and confirmed by domain experts. Furthermore, the dataset was benchmarked using multiple generations of You Only Look Once (YOLO) object detection models (Baseline Validation), which have been demonstrated to handle both large and small object detection in high-resolution images. The evaluation results, including confusion matrices and t-distributed Stochastic Neighbor Embedding (t-SNE) feature clustering diagrams, demonstrate the reliability and effectiveness of this dataset for training high-precision fraud detection models. This dataset is intended to support computer vision and anti-fraud research, promoting the automated development of fuel dispenser tampering detection. Full article

This study investigates how YouTube food travel vloggers semiotically construct destination images and potentially function as informal culinary ambassadors through gastrodiplomacy mechanisms, using Uzbekistan as a case study of emerging tourism markets. Although digital content creators are increasingly recognised as shaping tourism flows, a systematic understanding of the multimodal semiotic mechanisms through which food travel vlogs construct destination meanings remains limited. Using multimodal discourse analysis, this study examines six YouTube food travel videos on Uzbekistan (over 28 million combined views) from two prominent creators. The analysis integrates Kress and van Leeuwen’s visual grammar, Halliday’s systemic functional linguistics, van Leeuwen’s sound semiotics, and Norris’s multimodal interaction analysis to code a 60-segment corpus. Comparative analysis reveals 25 notable differences in semiotic features between the two creators, identifying two distinct semiotic profiles. Vlogger 1 primarily follows a parasocial intimacy model marked by direct gaze (89.2%), frequent second-person address (78.4%), and comparatively minimal editing. In contrast, Vlogger 2 adopts a cinematic documentary model characterised by first-person narration (56.5%), constructed visuals (60.9%), and gastronomic heritage narratives (34.8%). Despite these divergences, shared conventions centred on food composition, upbeat music, positive evaluation, and sharing gestures indicate a stable semiotic grammar of food travel vlogging. Analysis further reveals that orientalist dynamics and resistance to orientalism coexist within the same representational practice phenomenon termed ‘layered orientalism’, with distinct implications for how emerging destinations are mediated to international audiences. These findings suggest that digital content creators may employ distinct semiotic strategies that could function as informal culinary ambassadors through gastrodiplomacy mechanisms, potentially constructing destination awareness and cultural meaning for international audiences. This study contributes to theory on multimodal destination image construction and offers implications for how emerging tourism destinations might leverage multi-creator strategies to build culturally grounded destination brands. Full article

To accurately identify characteristics of the tourist experience, optimize tourism management and shape urban tourism brands, this study uses Wuhan as a case and aggregates multimodal user-generated content (UGC) data including tourist reviews, photos and travel vlogs. Based on the “Behavior–Cognition–Affect” framework and the progressive “Region–Route–Site” spatial perspective, this study adopts spatial analysis, image analysis, semantic network analysis, and natural language processing (NLP) to examine tourists’ spatial behavior patterns, visual cognitive preferences, and emotional feedback across urban, attraction, and individual tourist scales. Results show that Wuhan’s tourism presents a “core-periphery” spatial structure, tourists’ visual focus differs significantly across scenic types, and tourists’ emotions are generally positive, with consumption, shopping, and transportation as main negative sources. This study enriches the application of multimodal UGC in tourism geography, providing data to optimize tourism resource allocation and shape urban tourism images. Full article

Building reliable vehicle detection models for intelligent transportation systems calls for large, well-annotated datasets—yet gathering and labelling real traffic data remains both costly and labour-intensive. This paper introduces Traffic Synth, an automated pipeline that generates synthetic training datasets by altering real traffic camera images rather than constructing entirely artificial scenes. The system begins by detecting vehicles through instance segmentation and removing them from the frame. It then places new vehicles directly into the cleared regions using diffusion-based inpainting, all while retaining the original road layout, lighting, and camera perspective. Doing so preserves the realistic scene context while broadening the visual variety of vehicles in the dataset. To ensure that the resulting traffic looks physically plausible, we incorporate a lane-aware prompting mechanism that matches each vehicle’s orientation to the direction of travel as seen from the camera. The system further draws on a weighted vehicle brand database that mirrors the makes and colours commonly found on European roads to better match actual deployment conditions. Class-specific mask processing—involving anisotropic scaling and relative dilation—rounds out the pipeline by improving generation quality across different vehicle size categories. The final output is a set of images with automatically generated annotations in a standard object detection format. Full article

Disparity estimation from stereo-endoscopic images is critical for 3D reconstruction in minimally invasive surgery (MIS). However, surgical environments have inherent interference factors including soft tissue deformation, motion blur, and photometric inconsistency. Currently, self-supervised generative networks such as StyleGAN offer an alternative method, but their reliance on iterative latent optimization leads to high computational latency and limits practical deployment. In this work, we propose a temporal latent prediction method to accelerate this optimization process. Instead of designing a brand new generator, our framework learns to predict an optimized initial latent vector, thereby reducing the number of optimization steps and per-frame inference time. Crucially, this prediction-guided mechanism does not alter the architecture or inference logic of the generator, ensuring the fidelity of reconstruction is comparable to that of the original method. Experiments on Phantom and In vivo datasets demonstrate that our method reduces average optimization steps by 16–59% and cuts per-frame latency by about 2.3×, compared to baseline predictors and initialization strategies. Importantly, the final photometric loss remains nearly identical across all methods, confirming that acceleration does not compromise reconstruction quality. These results position our approach as a practical step toward efficient, self-supervised stereo-endoscopic reconstruction in clinical settings. Full article

A method for integrated brand analysis and strategy is developed in this work. The foundation of this method is market research, through which the relevance of brand attributes, their evaluation for competing brands and the market performance of these brands on the steps of the buying process are determined. On this basis, the overall evaluation of brands and their number of brand attributes with the best evaluation are calculated so that strategic decision guidelines for overall brand positioning can be deduced. These strategic decision guidelines are securing the brand based on the existing identity/image, developing the brand based on the existing identity/image, developing (pivoting to) a new brand identity/image, whilst securing the strengths of the existing identity/image, and developing a new brand identity/image. On the level of brand attributes, the weighted relevance of attributes and their evaluation difference to the best competitor are calculated so that, again, strategic decision guidelines can be deduced. The strategic decision guidelines on brand attribute level are securing the attributes as the core brand identity (first priority), selecting and developing the attributes to the core brand identity (second priority), securing the attributes as the extended brand identity (third priority), and selecting and developing the attributes as the extended brand identity (fourth priority). Based on the market performance of brands across the stages of the buying process, the conversions between these steps are determined. On this basis, strategic decision guidelines for market cultivation are deduced, i.e., awareness, image, sales, and loyalty strategies. To gain first indications of the validity of the method for integrated brand analysis and strategy, it is applied to food retail and chocolate brands in the German market. Future research should focus on further validating the method and enhancing it by integrating segmenting and targeting processes and, potentially, marketing measures on an operational level. Full article