Evaluation of facial changes after manual or movement-based interventions remains difficult because small and localized changes are often assessed using photographs or subjective impressions. This exploratory study aimed to visualize relative facial displacement before and after a low-load esthetic dance intervention using stereo ultraviolet digital image correlation (stereo UV-DIC). Twelve healthy adults performed an approximately 3 min intervention consisting of light whole-body rhythmic movements and facial self-massage focused mainly on the cheek region. Stereo ultraviolet facial images were acquired before and after the intervention under neutral facial expression and elevated-mouth-corner conditions. Melanin pigments visualized under ultraviolet-A illumination served as intrinsic random patterns for stereo UV-DIC. After three-dimensional facial surface reconstruction and rigid-body-motion correction, the directionality of relative displacement vectors was evaluated. Subjective mood, fatigue, body temperature, blood pressure, and pulse rate were also measured. Relative facial displacement was visualized in all participants. Although substantial interparticipant variability was observed, several participants showed upward or outward displacement tendencies, particularly under the neutral facial expression condition. Subjective mood and fatigue showed trends toward improvement, whereas vital signs showed no consistent marked changes. These findings suggest that stereo UV-DIC may be useful for non-contact exploratory evaluation of local facial displacement associated with low-load esthetic dance, while no consistent marked changes were observed in the measured vital signs.
Mixing across a stable density interface is inhibited by buoyancy. Two successively generated vortex rings provide localized, repeatable forcing, but the effect of the second ring depends on the transient interface and vortical field left by the first. Our previous experiment (Cao et al. 2021) showed that interfacial mixing varies with the emission interval, although planar measurements could not resolve the three-dimensional mechanism. The present study numerically determines how the emission interval controls this mechanism. Two vortex rings, denoted by VR_A and VR_B , are generated by a canonical piston-nozzle system and launched downward toward a sharp density interface. The nondimensional time and emission interval are defined as t^*=tu_0/d and Δt^*=Δtu_0/d , respectively, where u_0 and d are the characteristic piston velocity and nozzle diameter. At fixed circulation-based Reynolds number and Richardson number, which characterize vortex-induced advection relative to viscous diffusion and buoyancy relative to inertia, respectively, Δt^*=6.6 , 9.6, and 12.6 are examined. These intervals cause VR_B to encounter different stages of the interface rebound and interface-generated vortical field. The Δt^*=9.6 simulation reproduces the principal experimental features. The short interval produces an early, compact centerline interaction; the intermediate interval produces the strongest interface-centered deformation; and the long interval produces a delayed two-stage response. Within t^*≤ 36, the accumulated mixing ability is ordered as 6.6 > 9.6 > 12.6 because the contribution of VR_B begins earliest for the shortest interval. The results clarify how forcing timing controls the location, development, and accumulation of interfacial mixing and provide mechanistic guidance for selecting the timing of successive vortex-based forcing, rather than a universal optimum interval for practical systems.
This study experimentally investigated the translational and rotational dynamics of circular thin plates in a horizontal airflow using high-speed imaging technology. The motion parameters were extracted through contour recognition of the circular thin plates. Besides, the effects of incident angle, Reynolds number, and density ratio on the dynamic behavior were systematically examined. The results demonstrated that the motion of the circular plates could be categorized into three distinct regimes based on the maximum deflection angle: stable falling, deflected falling, and tumbling falling. Under the same flow conditions, smaller initial angle of attack could lead to higher dimensionless horizontal and vertical velocities as well as more intense rotational motion, accompanied by a faster and stronger angular response. The influence of Reynolds number was primarily reflected in the vertical velocity evolution, while its effect on overall rotational behavior was relatively limited. In addition, density ratio played a significant role in governing the dynamic response. Plates with lower density ratios exhibited enhanced sensitivity to airflow, characterized by larger translational velocities, earlier occurrence of peak angular velocity, and higher rotational intensity.
This study aimed to systematically map the global knowledge structure of intelligent sports research from 2010 to 2023. Specifically, it was designed to analyze the temporal evolution of publication output, identify major contributors and collaboration patterns, detect core research themes and emerging research directions, and generate structured insights into the developmental trajectory and international landscape of the field. We collected relevant literature from the Web of Science core database covering 2010–2023. Using VOSviewer software, we created knowledge maps of authors, institutions, countries, and keywords associated with intelligent sports-related literature. Additionally, we utilized CiteSpace software for keyword clustering mapping and Scimago Graphica to display a world map of publication volumes by country. 3044 articles related to intelligent sports were identified, showing a yearly increasing trend in publication volume. The results showed that China had the highest number of publications, with high-output research institutions being higher education establishments. Some institutions showed strong collaboration links within the retrieved dataset. Keyword co-occurrence and burst analyses indicated that major research themes included Deep and Machine Learning, Decision Making, Virtual and IoT Technologies, and Sports Vision, while keyword burst analysis identified emerging research directions involving Intelligent Sports Systems, Convolutional Neural Networks, and Devices. Over the past 14 years, the global publication volume on intelligent sports has steadily increased, though publication outputs and collaboration patterns vary across countries. Future research may benefit from more interdisciplinary and application-oriented approaches based on identified research themes and emerging directions. Multi-regional collaboration may provide opportunities to strengthen research connections and knowledge exchange in global intelligent sports research.
The aesthetic power of Chinese calligraphy is rooted in Qi Yun ((sic)(sic)), the expressive vitality and spiritual resonance conveyed through calligraphic forms, a quality that emerges from the multi-scale interplay of form, void, and rhythm across the entire composition rather than from any individual stroke. Despite centuries of rich theoretical discourse, this holistic spatial dimension has largely resisted objective analysis, leaving digital heritage scholarship reliant on qualitative connoisseurship. This paper presents a four-level visual analytics framework that quantifies the spatial structure of Chinese calligraphy from the macro to the micro-scale, spanning the overall composition, column, character, and stroke levels. Each level contributes interpretable computational metrics (brightness variation, boundary point nearest-neighbor distances, Voronoi spatial axes, inter-character spacing variance, center-of-mass dynamics, and stroke curvature entropy), all formally linked to named principles from classical calligraphy treatises. We evaluate the framework through a comparative case study on three structurally distinct versions of Mi Fu's Sui Feng Tie and through three extended quantitative experiments on spatial rhythm, structural dynamics, and kinematic energy. Results show that the framework differentiates the spatial and structural profiles of the historical reference, a seal-altered version, and a modern rewrite, and that each metric can be interpreted in relation to a corresponding classical aesthetic concept. Beyond cultural-heritage documentation, the framework offers practical tools for calligraphic education and suggests interpretable aesthetic constraints as a direction for generative AI systems.Graphical abstract Overview of the four-level visual analytics pipeline for Chinese calligraphy. Left column: preprocessing stages, original work, binary image, and contour polygon. Right panel: the four hierarchical analysis levels (top to bottom, by decreasing spatial scale)
