Thomas Baltzer Moeslund

Action spotting aims to localize actions in long, untrimmed videos using a single keyframe per action, making it well suited to the fast-paced nature of sports. Focusing on football, we identify two remaining challenges. First, although existing methods account for different temporal context lengths and scales, their relative effectiveness has not been systematically compared. Second, most existing methods rely exclusively on video features, requiring models to infer football-specific spatial and relational structure implicitly from appearance and motion, which is particularly challenging when only a limited number of annotated matches are available. We address these limitations through two contributions. First, we conduct a comparative analysis of single- and multi-scale temporal architectures, showing that multi-scale models provide only modest gains over single-scale alternatives in video-only settings relative to their additional computational cost. Second, we introduce TraVIS, an action spotting framework that complements video features with explicit player and ball tracking information. TraVIS represents the game state as a graph whose nodes encode players and the ball and whose edges capture their spatial and relational structure, thereby introducing football-specific inductive biases and explicitly representing information that video-only approaches must otherwise learn implicitly. Fusing the graph and video representations achieves the strongest performance among the evaluated configurations on a proprietary football dataset, improving over the video-only model by + 6.42 mAP@1s. These results show that structured tracking information provides a valuable complement to video features for football action spotting.

Background: Clinical documentation is crucial for maintaining continuity and quality of care; however, it has become increasingly time-consuming and complex in general practice, leading to administrative overload and burnout. Artificial intelligence-powered voice-to-text technologies (AIVT) have emerged as potential solutions to reduce this burden by automatically transcribing and organizing consultations into clinical notes. This study assesses perceptions of the utility of an AIVT tool for clinical documentation during video consultations. Methods: A qualitative study was conducted using triangulation of observations, think-aloud sessions, and semi-structured interviews. A total of 18 general practitioners from 14 clinics across all five Danish regions tested Autonotar, a use case for an AIVT tool that generates draft clinical notes during video consultations. Data were analyzed deductively using predefined criteria: demand, acceptability, and perceived future efficacy. Results: The findings indicate: (1) structural drivers of demand show how increasing regulatory and defensive documentation pressures create a perceived need for technological support; (2) conditions for acceptable use highlight where general practitioners emphasize the importance of clinical accuracy, alignment with workflow, and robust data protection; and (3) perceived future value demonstrates how participants view AIVT as a way to reduce administrative workload, improve work-life balance, and enhance patient engagement—provided the tool delivers reliable output and is responsibly integrated into clinical systems. Conclusions: Based on the study use case, AIVT may offer a practical means of supporting clinical documentation in Danish general practice, but its utility is conditional rather than inherent. GPs expressed cautious optimism about its potential to reduce documentation burden, improve work–life balance, and support patient engagement. However, adoption will depend on reliable output, workflow integration, clear accountability, robust data protection, and adaptation to different consultation modalities. Sustainable implementation requires technological maturity and sociotechnical alignment among clinicians, decision-makers, and technology providers.

In recent years, the underwater image formation model has been widely used to generate synthetic underwater data. Although many approaches focus on scenes primarily affected by discoloration, they often overlook the model’s ability to capture the complex, distance-dependent structural degradations present in highly turbid environments. This paper extends our previous work, published in the 2025 International Conference on Computer Vision (ICCV) workshop proceedings, in which we introduced STSR , an improved synthetic data generation pipeline that includes the commonly omitted forward-scattering term and accounts for a nonuniform medium. In this journal version, we significantly expand on the original study by providing a more thorough description of the methodology and implementation details, as well as an extensive quantitative evaluation of the resulting synthetic data. We find that images generated with STSR result in improved generalization ability across models and turbidity levels. Data and code can be accessed through the project page: https://vap.aau.dk/sea-ing-through-scattered-rays .

Despite rapid advances in artificial intelligence (AI), its adoption in Danish general practice remains limited and decentralized, relying on individual general practitioners' (GPs') decisions. This study aimed to evaluate Danish GPs' acceptance of AI and the importance of factors derived from established technology-acceptance models. A cross-sectional web survey was conducted from mid-September 2024 to March 2025, using a culturally adapted version of a previously validated tool. The survey included 42 items: nine background items and 33 items covering 11 factors, including Medical and Non-Medical Performance Expectancy, Effort Expectancy, Social Influence (Medical and Patient), Facilitating Conditions, Perceived Trust, Anxiety, Professional Identity, Innovativeness and Behavioral Intention. Descriptive statistics, subgroup comparisons and Cronbach's alpha were provided. A total of 109 GPs and residents responded, with 84.4% (n = 92) completing the survey. Attitudes toward AI were generally neutral to positive, with mean scores above neutral in seven of the 11 factors. The most positive factors measured were Medical and Non-Medical Performance Expectancy. Scores related to Behavioral Intention were also high. Lower scores appeared in Perceived Trust, Facilitating Conditions and Anxiety. Older GPs (60 years and above) reported lower scores on Non-Medical Performance Expectancy and Social Influence (Medical); GPs in smaller cities showed more positive attitudes across several factors. Cronbach's alpha indicated good internal consistency on all but two scales. In summary, Danish GPs demonstrate a strong intention to adopt AI. However, issues persist as significant barriers regarding Anxiety, Perceived Trust and Facilitating Conditions. Implementation strategies considering context and clinician demographics are recommended.