Personally curated content in short-form video formats provides added value for participants and spectators but is often disregarded in lower-level events because it is too labor-intensive to create or is not recorded at all. Our smart sensor-driven tripod focuses on supplying a unified sensor and video solution to capture personalized highlights for participants in various sporting events with low computational and hardware costs. The relevant parts of the video for each participant are automatically determined by using the timestamps of his/her received sensor data. This is achieved through a customizable clipping mechanism that processes and optimizes both video and sensor data. The clipping mechanism is driven by sensing nearby signals of Adaptive Network Topology (ANT+) capable devices worn by the athletes that provide both locality information and identification. The device was deployed and tested in an amateur-level cycling race in which it provided clips with a detection rate of 92.9%. The associated sensor data were used to automatically extract peloton passages and report riders’ positions on the course, as well as which participants were grouped together. Insights derived from sensor signals can be processed and published in real time, and an upload optimization scheme is proposed that can provide video clips for each rider a maximum of 5 min after the passage if video upload is enabled.
This paper explores the use of wearable technology (Garmin Fenix 7) to monitor physiological and psychological factors contributing to attrition during basic military training. Attrition, or the voluntary departure of recruits from the military, often results from physical and psychological challenges, such as fatigue, injury, and stress, which lead to significant costs for the military. To better understand and mitigate attrition, we designed and implemented a comprehensive and continuous data-capturing methodology to monitor 63 recruits during their basic infantry training. It’s optimized for military use by being minimally invasive (for both recruits and operators), preventing data leakage, and being built for scale. We analysed data collected from two test phases, focusing on seven key psychometric and physical features derived from baseline questionnaires and physiological measurements from wearable devices. The preliminary results revealed that recruits at risk of attrition tend to cluster in specific areas of the feature space in both Linear Discriminant Analysis (LDA) and Principal Component Analysis (PCA). Key indicators of attrition included low motivation, low resilience, and a stress mindset. Furthermore, we developed a predictive model using physiological data, such as sleep scores and step counts from Garmin devices, achieving a macro mean absolute error (MAE) of 0.74. This model suggests the potential to reduce the burden of daily wellness questionnaires by relying on continuous, unobtrusive monitoring.
Road cycling races, such as the Tour de France, captivate millions of viewers globally, combining competitive sportsmanship with the promotion of regional landmarks. Traditionally, points of interest (POIs) are highlighted during broadcasts using manually created static overlays, a process that is both outdated and labor-intensive. This paper presents a novel, fully automated methodology for detecting and tracking POIs in live helicopter video streams, aiming to streamline the visualization workflow and enhance viewer engagement. Our approach integrates a saliency and Segment Anything-based technique to propose potential POI regions, which are then recognized using a keypoint matching method that requires only a few reference images. This system supports both automatic and semi-automatic operations, allowing video editors to intervene when necessary, thereby balancing automation with manual control. The proposed pipeline demonstrated high effectiveness, achieving over 75% precision and recall in POI detection, and offers two tracking solutions: a traditional MedianFlow tracker and an advanced SAM 2 tracker. While the former provides speed and simplicity, the latter delivers superior segmentation tracking, albeit with higher computational demands. Our findings suggest that this methodology significantly reduces manual workload and opens new possibilities for interactive visualizations, enhancing the live viewing experience of cycling races.
Despite volleyball not being a contact sport, the sport's high-intensity movements - such as jumps, dives, and sprints - pose injury risks to players. In recent years, the surge in computer vision applications within sports has offered robust analysis tools for both coaches and players. However, the existing applications in volleyball still lack comprehensive 3D-based metrics in a non-laboratory setting. This work addresses this gap by developing a semi-automated algorithm specifically designed to reconstruct volleyball players in 3D, using a calibrated multi-camera set-up able to capture synchronized videos. The primary goal of this research is to equip coaches with a valuable tool for assessing individual players' physical loads enabling them to make more informed decisions or adjust their training program to prevent injuries. Detailed analyses of player jumps, dives, and trajectories within rallies provide insights into the algorithm's performance and practical application in volleyball settings. Our findings reveal promising advancements in jump and dive detection, as well as jump height estimation, as our estimates are closer to the ground truth than those of the widely employed Vert sensors. Moreover, our findings highlight the significance of a 3D player reconstruction algorithm, concluding that the implications go far beyond the explored analyses in this work as there are numerous other potentially intriguing avenues left unexplored.