Motion History Research Articles

Time-weighted motion history image for human activity classification in sports

Vision-based human activity classification has remarkable potential for various applications in the sports context (e.g., motion analysis for performance enhancement, active sensing for athletes, etc.). Recently, learning-based human activity classifications have been widely researched. However, in sports scenes in which more detailed and player-specific classifications are required, this is a quite challenging task; in many cases, only a limited number of datasets are available, unlike daily movements such as walking or climbing stairs. Therefore, this paper proposes a time-weighted motion history image, an effective image sequence representation for learning-based human activity classification. Unlike conventional MHI based on the assumption that “the newer frame is more important,” our method generates importance-aware representation so that the predictor can “see” the frames that contribute to analyzing the specific human activity. Experimental results have shown the superiority of our method.

Hydrodynamic simulations of the KT Eridani nova super-remnant

ABSTRACT A nova super-remnant (NSR) is an immense structure associated with a nova that forms when frequent and recurrent nova (RN) eruptions sweep up surrounding interstellar medium (ISM) into a high-density and distant shell. The prototypical NSR, measuring over 100 pc across, was discovered in 2014 around the annually erupting nova M 31N 2008-12a. Hydrodynamical simulations demonstrated that the creation of a dynamic NSR by repeated eruptions transporting large quantities of ISM is not only feasible but that these structures should exist around all novae, whether the white dwarf (WD) is increasing or decreasing in mass. But it is only the RN with the highest WD masses and accretion rates that should host observable NSRs. KT Eridani is, potentially, the eleventh RNe recorded in the Galaxy and is also surrounded by a recently unveiled H α shell tens of parsecs across, consistent with an NSR. Through modelling the nova ejecta from KT Eri, we demonstrate that such an observable NSR could form in approximately 50 000 yr, which fits with the proper motion history of the nova. We compute the expected H α emission from the KT Eri NSR and predict that the structure might be accessible to wide-field X-ray facilities.

Rapid damage assessment of regional beam bridges after earthquake based on simplified models and different seismic inputs

As the hub of the traffic network, the rapid damage assessment of bridges after earthquake is of great significance for post-disaster rescue. Traditional vulnerability-based method can only provide the probability of different damage states of structures under different earthquake intensity indicators, which confuses the selection of post-disaster rescue routes. Therefore, this paper proposes a post earthquake damage assessment technique for regional bridges based on nonlinear dynamic time-history response analysis. First, a simplified mechanical model for a single pier was established for a large number of beam bridges in the region, achieving standardized and batch rapid finite element model establishment. Secondly, using Kriging interpolation method and continuous wavelet transform technology, the surface seismic response of geological drilling points is calculated to obtain the seismic motion history of any bridge site in the region. Finally, using parallel computing technology, nonlinear dynamic time-history response analysis was conducted on the simplified beam bridge model, and the bridge damage status was quantitatively evaluated based on the calculation results. Eighty-five bridges in a specific area are considered as an example to verify the proposed method. The results show that the proposed method has high calculation efficiency and can consider the site characteristics and structural design parameters. Also, the calculation results can better reflect the actual earthquake damage distribution, providing strong technical support for post-disaster rescue.

Non-local tails in radiation in odd dimensions

Huygens principle violation in a spacetime of odd dimensions leads to the fact that the retarded massless fields of localised sources depend on their history of motion preceding the retarded time. This non-local character of retarded fields should result into the formation of tail signals in the radiation of localised sources. In particular, in gravity theories with odd number of extra spacetime dimensions the gravitational radiation of binary systems should contain the tail terms. In this work, we demonstrate the presence of tail signal in radiation within a simple model of scalar field interacting with the point charge moving on elliptical orbit in three dimensions. We find that the tail term results into the characteristic dependence of radiation power of the charge on time. In particular, its extremum points do not correspond to the moments when the charge passes the pericenter and apocenter of the orbit, in contrast with the four-dimensional theory. We obtain the formulae for the shifts of radiation power extremum points up to the contributions quadratic in the orbital eccentricity. We also compute the spectral distribution of radiation power of the charge. We find that in three dimensions the charge on elliptical orbit radiates into the lower harmonics of the spectrum, compared to the four-dimensional theory. We conjecture that in higher dimensions the character of spectral distributions is opposite — the charge mainly radiates into the higher harmonics of the spectrum.

Comparison of physics-based and machine learning methods for phase-resolved prediction of waves measured in the field

Phase-resolved predictions of surface waves can be used to optimize a wide variety of marine applications. In this paper, we compare predictions obtained using two independent methods for field data, with horizons sufficient to control wave energy converters. The first method is physics-based prediction. In this method, a set of optimal representative angles, obtained using an optimization algorithm given time histories of a wave buoy motion in 3D, are used for forward propagation based on linear wave theory. The second method is a machine learning method using an Artificial Neural Network (ANN) which requires longer records for training. Field measurements were obtained from the Southern Ocean of Albany, WA. The field data were collected by an upwave ‘detection’ array of 3 Sofar Spotter wave buoys and a downwave ‘prediction’ point coincident with a Datawell Waverider-4. All buoys were soft-moored, and data were collected over 3 months in 2022. Selected intervals during this period are presented in the paper to compare and contrast the predictions made by the two different methods. We find that some wave fields can be predicted well over more than a period in advance, all that is required for active control of a wave power take-off in a renewable energy application. In contrast, highly spread sea states remain a challenge. The methods are also compared in terms of the complexity and time required for making predictions. Further discussions are made on the applicability of the results to other locations.

Forearc spreading cessation and backarc basin evolution: Insights from Mariana subduction system and geodynamic modeling

Evolutions of forearc and backarc spreading are related to processes of Subduction Initiation (SI). The unique evolution history of the Izu-Bonin-Mariana subduction system (IBM) provides a perfect example to investigate the relationships between forearc and backarc spreading with SI. The plate reconstruction model shows that the IBM initiated around 52 Ma, and the kinematics of the Pacific Plate changed around 47 Ma, providing horizontal velocity for forearc spreading cessation and backarc opening. Here, based on the plate motion history and the 2D thermomechanical models, we simulated the early stages of SI, as well as the evolutions of forearc and backarc spreading. We proposed a new model with a delayed convergence rate after ∼5 Myr of SI, which can explain 1) the observed lithospheric forearc spreading cessation of the IBM; 2) the observed formation time of the Shikoku basin; and 3) the horizontally lying slab in the mantle transition zone after cessation of the backarc spreading of the IBM.

Self-force of high-speed dislocation in anisotropic media based on configurational mechanics

Subsonic, transonic, and supersonic dislocations play a critical role during super high strain rate (>105/s) deformation. For these high-speed dislocations, the self-force is history dependent, which is different from and much more complex than the quasi-static dislocations. Solutions to the self-force of high-speed dislocation in continuum media are only known for some special cases, i.e., subsonic dislocation moving in isotropic infinite crystals with a constant speed. However, metals are generally anisotropic, and high-speed dislocations may have complicated histories and interact with the free surface. How crystal anisotropy, complicated motion history and free surface influence their self-force has so far received little attention. In the current work, we proposed an effective calculation method of self-force on high-speed dislocation based on the discrete-continuous model of three-dimensional dislocation elastodynamics and the dynamic J-integral of configurational mechanics. It is applicable to arbitrarily moving subsonic, transonic and supersonic dislocation, in both isotropic and anisotropic crystals, and can automatically consider the image force if the dislocation is close to the free surface. The effectiveness of the method is carefully verified by comparing it with the existing theoretical solutions and the molecular dynamics results. The effect of crystal anisotropy on the elastodynamic stress field, as well as the self-force of high-speed dislocation close to or far from free surface are studied.

Effect of the supercontinent cycle on the longest-term sea-level change from a simple conceptual and theoretical model

The longest-term (first-order) sea-level change is considered to be within ∼ 200–300 million years, which is roughly half or less of the period of the supercontinent cycle of ∼ 500–700 Myr. It is recognized that the assembly of supercontinents involves either the closure of the “interior ocean” that resulted from the breakup of the supercontinent or the closure of the “exterior ocean” that surrounded the supercontinent. The former process is termed “extroversion,” while the latter is termed “introversion,” both of which are an end-member scenario of the supercontinent cycle. A “combination” process is also proposed to explain the complex behavior of continental drift on the actual Earth. Despite recent advances in a time-dependent mantle convection model, some difficulties still remain in self-consistently recovering plate tectonics and continental drift of the past Earth. Therefore, it is worth addressing the fundamental physics of the effects of the history of plate motion and the associated continental drift on the longest-term sea-level change from the perspevtive of the supercontinent cycle in a timescale of the order of 100 million years. In this study, a simple conceptual model of mantle convection with various supercontinent cycle processes was constructed, and the effects of the supercontinent cycle on the longest-term sea-level change were addressed. The sea-level change is estimated from bathymetry over the oceanic plates under different free parameters that control the fluctuation of the sea-level change. The present analytical results suggest that the asymptotic bathymetry due to the plate flattening effect of oceanic plates in the exterior and interior oceans is essential to recover the realistic magnitude of sea-level fluctuations that is comparable to the sea-level curve of the past Earth. In addition, the absolute volume change in the seawater due to its recycling into the mantle may be important to fit the observed sea-level curves from stratigraphic studies since 200 million years ago.

Non-parameterized ship maneuvering model of Deep Neural Networks based on real voyage data-driven

While Deep Neural Network (DNN) models for ship maneuvering model are commonly constructed using experimental model ships or simulation data, this study focuses on training four Non-parameter Ship maneuvering Network models (NSM_Net_models) by using real ship voyage data, then evaluating their performance by comparison. To support the data-driven approach, real ship voyage data is collected from the Voyage Data Recorder (VDR) system of a training ship. Along with this, the input variables of the models are enhanced by incorporating the past movement history of ships. Furthermore, the impact of three different durations of past ship motion history on the models’ performance is compared and analyzed. The results suggest the feasibility and practical application value of the ship maneuvering motion model built by the deep network.

A-15 Predictive Utility of Risk Factors Associated with Vestibular and Oculomotor Outcomes Following Sport-Related Concussion

Abstract Purpose The purpose of the present study was to determine the ability of these risk factors to predict vestibular/oculomotor symptom provocation on the Vestibular/Oculomotor Screening (VOMS). It was hypothesized that both pre- and post-injury risk factors would predict symptom provocation on the VOMS. Methods This was a retrospective chart review of patients (n=78; 41 male) aged 15.2±1.7 years (range 12-18 years) seeking care at a concussion specialty clinic. Participants were seen 5.1±3.2 days (range 1-14 days) post-injury and completed a clinical interview, history questionnaire, symptom inventory, and VOMS as part of routine clinical care. Data analysis included chi-square tests with odds ratios to examine the association between identified risk factors and VOMS outcomes. An alpha value of .05 was used for all analyses. Results Personal history of motion sickness (χ2=7.6, p=.006, OR=11.4), personal history of oculomotor dysfunction (Fisher’s exact test, p=.01, OR=1.2), post-traumatic migraine (PTM) symptoms (χ2=11.7, p<.001, OR=5.4), and post-injury fogginess (χ2=10.1, p=.001, OR=4.8), were associated with presence of one or more post-injury VOMS individual item scores above clinical cutoff (i.e., symptom elevation >2). History of oculomotor dysfunction (Fisher’s exact test p=.013, OR=6.6) and PTM symptoms (χ2=5.0, p=.025, OR=3.3) were associated with obtaining a post-injury near-point convergence (NPC) distance >5 cm. Gender, loss of consciousness status, concussion history, history of migraine, and psychiatric history were not associated with VOMS symptom provocation or abnormal NPC. Conclusions Findings indicate that both pre-injury risk factors (i.e., personal history of motion sickness and history of oculomotor dysfunction) and post-injury risk factors (i.e., fogginess, PTM symptoms) were associated with clinically meaningful symptom provocation on the VOMS. In addition, history of oculomotor dysfunction and PTM symptoms were both associated with abnormal NPC distance. These results reinforce the importance of a comprehensive evaluation including a clinical history and vestibular and oculomotor screening following SRC. Consistent with prior research, loss of consciousness status and concussion history were not associated with increased risk for vestibular/oculomotor outcomes.

Early single sport specialization does not improve pitching velocity, motion, strength, or utilization in high school baseball pitchers

The impact of single-sport specialization on performance and injury risk in youth and adolescent athletes remains debated. The purpose of this study was to determine the incidence of single-sport specialization and its association with pitch velocity, pitcher utilization, range of motion, strength, and injury history. High school pitchers in the Northeastern, Southeastern, Midwestern, and Western Regions of the United States were prospectively recruited. Prior to the season, pitchers were surveyed about their basic demographics, sport participation history, and injury history. At this timepoint, a standardized physical examination was performed and max pitch velocity was measured. They were then prospectively followed over the ensuing high school season, where pitch utilization and counts were followed and injuries were tracked. Across the study, 115 pitchers were recruited, of whom 53% played baseball as their only sport and 47% played other sports. Single- position specialization within baseball was uncommon as 90% of athletes played positions other than pitcher. There were no differences in range of motion, strength, or pitch velocity between multisport athletes and single-sport specialists. Within the multisport group, 26% reported a history of injury, while 38% of the single-sport specialists reported a history of injury (P= .179). There were no significant differences in pitcher utilization between multisport athletes and single-sport specialists. Only a single prospective injury occurred in the 115 players. Although single-sport specialization is common, nearly half of all high school pitchers in this study were multisport athletes. While injury history was generally similar between the two groups, single-sport baseball players did not have the following: improved motion, increased strength, higher ball velocity, or greater pitching volumes during the high school baseball season compared to multisport athletes. Accordingly, single-sport specialization did not have a competitive advantage for these high school pitchers based on the performance factors studied. Further prospective studies are necessary to understand the association between single-sport specialization and injury risk in high school baseball players.

Human Behavior Recognition via Hierarchical Patches Descriptor and Approximate Locality-Constrained Linear Coding.

Human behavior recognition technology is widely adopted in intelligent surveillance, human-machine interaction, video retrieval, and ambient intelligence applications. To achieve efficient and accurate human behavior recognition, a unique approach based on the hierarchical patches descriptor (HPD) and approximate locality-constrained linear coding (ALLC) algorithm is proposed. The HPD is a detailed local feature description, and ALLC is a fast coding method, which makes it more computationally efficient than some competitive feature-coding methods. Firstly, energy image species were calculated to describe human behavior in a global manner. Secondly, an HPD was constructed to describe human behaviors in detail through the spatial pyramid matching method. Finally, ALLC was employed to encode the patches of each level, and a feature coding with good structural characteristics and local sparsity smoothness was obtained for recognition. The recognition experimental results on both Weizmann and DHA datasets demonstrated that the accuracy of five energy image species combined with HPD and ALLC was relatively high, scoring 100% in motion history image (MHI), 98.77% in motion energy image (MEI), 93.28% in average motion energy image (AMEI), 94.68% in enhanced motion energy image (EMEI), and 95.62% in motion entropy image (MEnI).

Multi-fidelity nonlinear unsteady aerodynamic modeling and uncertainty estimation based on Hierarchical Kriging

By fusing aerodynamic data from multiple sources, multi-fidelity methods can well balance model accuracy and computational cost. To extend multi-fidelity models for predicting unsteady aerodynamics with uncertainty estimation, an improved modelling framework based on the Hierarchical Kriging (HK) is proposed for nonlinear aerodynamic reduced-order modelling. This aerodynamic modelling framework is achieved by a two-stage modelling strategy. In the first stage, using the current and time-delayed low-fidelity aerodynamic load and pitching motion as input, a linear regression model is built to achieve preliminary aerodynamic prediction. In the second stage, the current and time-delayed preliminary prediction, as well as pitching motion history are used as the input of the Hierarchical Kriging model to predict the high-fidelity aerodynamic loads. The generalization capability of this improved modelling framework is firstly verified by several analytical examples. Furthermore, with three sets of training samples taken from harmonic pitching motions of an airfoil in transonic flow, the proposed model achieves accurate prediction on unsteady aerodynamics of other harmonic and random motions, and provides reliable uncertainty estimation of the predicted unsteady aerodynamics. In addition, the computational cost and prediction errors of multi-fidelity and the single-fidelity aerodynamic model are compared. Even with the same computational cost, the proposed multi-fidelity model still shows improved accuracy. Finally, using the uncertainty estimation of the proposed model, an active learning task is performed, which actively selects and incrementally adds the required training samples. With active learning, a good balance between model accuracy and cost of data acquisition is maintained, while avoiding over-fitting problems of the proposed model.

Deterministic Seismic Hazard Analysis to Determine Liquefaction Potential Due to Earthquake

The great rocking of building structures and the occurrence of liquefaction in water-saturated soil on river banks are generally caused by earthquake shaking. The waves generated by the earthquake are the main cause of the shaking. In order to show the effect of ground motion earthquake shaking on the response of structures and liquefaction processes, it is necessary to analyze the structure and liquefaction as well as the time history of artificial earthquake ground motions. An artificial time history for liquefaction analysis can be developed based on spectral matching to the target spectrum generated by a deterministic seismic hazard analysis. Therefore, the time history recovered from the analysis can be said to be derived from a deterministic procedure. The analysis of liquefaction with time history aims to see the potential for liquefaction in the Palu region of Central Sulawesi by developing the time history of the bedrock. The time history of the bedrock is then spread over the ground surface. The propagation of time-historical waves to the ground surface can cause liquefaction events in the soil layer. It was found that liquefaction occurred in the Palu region, especially in the Anutapura Hospital building. No other liquefaction potential analysis studies were found in the region. Doi: 10.28991/CEJ-2023-09-05-012 Full Text: PDF

Nature and preservation of Late Jurassic breakup-related volcanism in the Carnarvon Basin, North West Shelf, Australia

The North West Australian Margin, which formed as Greater India rifted from Australia during the Jurassic to Early Cretaceous, is recognised as an archetypal magma-rich rifted margin, with records of extensive igneous activity in the Exmouth Plateau and Exmouth Sub-Basin (ESB) of the Northern Carnarvon Basin (NCB). Pre-breakup magmatism is manifested by a large ∼400 × 150 km intrusive sill complex, emplaced into Triassic and Jurassic strata in the Late Jurassic and Early Cretaceous. An apparent lack of extrusive igneous rocks has caused previous works to describe the region as a large intrusive igneous province. Here, we describe two recently identified Upper Jurassic volcanic centres: the Pyrenees Volcano in the eastern ESB (first reported here), and the Toro Volcanic Complex (TVC), in the western ESB. Although offset by Early Cretaceous normal faulting, the edifice of the Pyrenees Volcano and associated lava flows are well preserved beneath a protective carapace of Upper Jurassic strata below the angular Intra-Hauterivian Unconformity on the Novara Arch. In contrast, a significant proportion of the TVC was peneplaned beneath an intra-Valanginian (Early Cretaceous) unconformity following breakup-related uplift. As Upper Triassic to Lower Cretaceous strata appear to have been eroded over the Ningaloo Arch in the southern ESB, we postulate that Late Jurassic extrusive volcanism may have been more spatially extensive, prior to erosion associated with Early Cretaceous exhumation in the southern NCB. Hence our findings suggest that the NCB was potentially host to significantly more extrusive volcanism than has been preserved within basin fill. Our findings also have broader implications for the conditions required to preserve extrusive igneous material in sedimentary basins within large igneous provinces that have undergone complex histories of rift-related vertical motion.

Scaling effects on the seismic response of a closed-end pipe pile embedded in dry and saturated coarse grain soils

Foundations can be subjected to dynamic or seismic loads depending on their applications and the site being constructed in. The researchers concentrated their works on investigating the reasons of the significant damage of piles during seismic excitation. Based on the findings of laboratory experiments and other numerical analyses, such failures were referred to as the kinematic impact of the earthquake on piles since they were associated with discontinuities in the subsoil because of sudden changes in soil stiffness. The current work investigates the seismic response of closed-end (CE) pipe pile using three-dimensional finite element analysis, including the impact of the scaling-up model, acceleration-time history of the ground motion, and ground conditions. The numerical model is developed using a variety of scaling rules and the outputs of the available laboratory tests. The current results showed that the saturated sand models have larger pile deformation factors than dry sand models. Pile frictional resistance was evaluated numerically, and the entire findings were evaluated against the earlier work. Mainly, the frictional resistance around the pile shaft was lower than that at the pile tip, and the frictional resistance factor on the soil surface of dry soil models was larger than that of saturated soil models. Owing to the acceleration amplifications, the pile and soil suffered cycles of compression and tension stresses. A hysteresis loop is broader and flatter on the x-axis as the shear strain increases serve to identify the shear stress–strain plane behavior. The main outputs of the scaled models were normalized to provide a deep insight of model to prototype scaling effects.

A Brief History of Motion: From the Wheel, to the Car, to What Comes Next

"A Brief History of Motion: From the Wheel, to the Car, to What Comes Next." Journal of the American Planning Association, ahead-of-print(ahead-of-print), p. 1

Weakly nonlinear dynamics of a chemically active particle near the threshold for spontaneous motion. II. History-dependent motion

Isotropic chemically active particles can spontaneously self-propel owing to a symmetry-breaking instability when diffusion of the chemical is sufficiently weak relative to advection. This series revisits the weakly nonlinear theory describing the dynamics of such spontaneous swimmers near the instability threshold. Part II presents an extension to the general framework from part I to include unsteadiness in the form of an integral over the history of the particle motion, representing the interaction of the particle with its own chemical wake. This allows efficient simulation and theoretical analysis of fully three-dimensional unsteady problems in a range of physical scenarios.

Semantic scene upgrades for trajectory prediction.

Understanding pedestrian motion is critical for many real-world applications, e.g., autonomous driving and social robot navigation. It is a challenging problem since autonomous agents require complete understanding of its surroundings including complex spatial, social and scene dependencies. In trajectory prediction research, spatial and social interactions are widely studied while scene understanding has received less attention. In this paper, we study the effectiveness of different encoding mechanisms to understand the influence of the scene on pedestrian trajectories. We leverage a recurrent Variational Autoencoder to encode pedestrian motion history, its social interaction with other pedestrians and semantic scene information. We then evaluate the performance on various public datasets, such as ETH-UCY, Stanford Drone and Grand Central Station. Experimental results show that utilizing a fully segmented map, for explicit scene semantics, out performs other variants of scene representations (semantic and CNN embedding) for trajectory prediction tasks.

Two-Stream Deep Learning Architecture-Based Human Action Recognition

Human action recognition (HAR) based on Artificial intelligence reasoning is the most important research area in computer vision. Big breakthroughs in this field have been observed in the last few years; additionally, the interest in research in this field is evolving, such as understanding of actions and scenes, studying human joints, and human posture recognition. Many HAR techniques are introduced in the literature. Nonetheless, the challenge of redundant and irrelevant features reduces recognition accuracy. They also faced a few other challenges, such as differing perspectives, environmental conditions, and temporal variations, among others. In this work, a deep learning and improved whale optimization algorithm based framework is proposed for HAR. The proposed framework consists of a few core stages i.e., frames initial preprocessing, fine-tuned pre-trained deep learning models through transfer learning (TL), features fusion using modified serial based approach, and improved whale optimization based best features selection for final classification. Two pre-trained deep learning models such as InceptionV3 and Resnet101 are fine-tuned and TL is employed to train on action recognition datasets. The fusion process increases the length of feature vectors; therefore, improved whale optimization algorithm is proposed and selects the best features. The best selected features are finally classified using machine learning (ML) classifiers. Four publicly accessible datasets such as Ut-interaction, Hollywood, Free Viewpoint Action Recognition using Motion History Volumes (IXMAS), and centre of computer vision (UCF) Sports, are employed and achieved the testing accuracy of 100%, 99.9%, 99.1%, and 100% respectively. Comparison with state of the art techniques (SOTA), the proposed method showed the improved accuracy.