Representation Of Motion Research Articles

Are Generations Really Divided by Climate? Preference for Conflict in Fridays for Future Media Coverage

ABSTRACT This paper analyses the role of the media in the representation of youth climate action, specifically the media tendency to construct the age-related aspects of climate change through a conflictual framing. A qualitative analysis of 347 articles on the Fridays for Future (FFF) activities produced by a diverse sample of the mainstream online Czech media reveals how media are generally absorbed by the generational perspective and conflict when covering the issue, but in two different ways reflecting the position towards the FFF. Conflictual framing is either based on a negative representation of the FFF movement itself, or, in the dominant case when the media are supportive of the movement, is created through an emphasis on the theme of intergenerational justice. We follow the theoretical approach of peace journalism calling for a substantial change in the media coverage of global crises and argue that, given the already polarised public opinion in Europe (and in the Czech Republic specifically), the media tendency to support yet another polarising perspective, based on age-differences this time, does not contribute to the possible solution of the problem.

Stochastic video normality network for abnormal event detection in surveillance videos

Video Anomaly Detection (VAD) aims to automatically identify unexpected spatial–temporal patterns to detect abnormal events in surveillance videos. Existing unsupervised VAD methods either use a single proxy task to represent the prototypical normal patterns, or split the video into static and dynamic parts and learn spatial and temporal normality separately, which cannot model the inherent uncertainty of motion. In this regard, we propose for the first time to learn video normality via stochastic motion representation. Specifically, the proposed Stochastic Video Normality (SVN) network learns the prototypical local appearance patterns via deterministic multi-task learning and global motion patterns in a non-deterministic manner. The stochastic representation module uses recurrent networks to model historical motion as a conditional Gaussian distribution and infers motion explicitly with the learned prior. Besides, a mask auto-encoder is introduced to explore the relationship between appearance and motion, boosting the model to learn spatial–temporal normality. Experimental results show that our proposed SVN network performs comparably to the state-of-the-art methods. Extensive analysis demonstrates the effectiveness of multi-task learning-based appearance representation and stochastic motion representation for unsupervised video anomaly detection.

Improved Representation of Groundwater–Surface Water Interactions Using SWAT+gwflow and Modifications to the gwflow Module

Recent water availability and scarcity problems have highlighted the importance of surface–groundwater interactions. Thus, groundwater models are coupled with surface water models. However, this solution is complex, needing code modifications and long computation times. Recently, a new groundwater module (gwflow) was developed directly inside the SWAT code to tackle these issues. This research assesses gwflow’s capabilities in representing surface–groundwater system interactions in the Dijle catchment (892.54 km2), a groundwater-driven watershed in Belgium. Additional developments were made in SWAT+gwflow to represent the interaction between the groundwater and soil (gwsoil). The model was calibrated for monthly mean streamflow at the catchment outlet (1983 to 1996) and validated for two periods (validation 1: 1975 to 1982 and validation 2: 1997 to 2002). It was found that the SWAT+gwflow model is better at representing the total flow (NSE of 0.6) than the standalone SWAT+ (NSE of 0.4). This was confirmed during two validation periods where the standalone model scored unsatisfactory monthly NSE (0.6 and 0.1), while the new model’s NSE was 0.7 and 0.5. Additionally, the SWAT+gwflow model simulations better depict the groundwater via baseflow and attain proper water balance values. Thus, in a highly groundwater-driven catchment, the simplified representation of groundwater systems by the standalone SWAT+ model has pitfalls. In addition, the modification made to the gwflow module (gwsoil) improved the model’s performance, which, without such adjustment, overestimates the streamflow via saturation excess flow. When including the gwsoil mechanism, thereby providing a more accurate representation of water storage and movement, groundwater is transferred to the soil profile, increasing the overall soil water content and thereby increasing lateral flow. This novel modification can also have implications for other distributed hydrological models to consider such exchanges in their modeling scheme.

יהדות ארצות הברית וייצוגה בטקסט החינוכי בבתי ספר ממלכתיים, 2003-1955

This article discusses the representation of American Jewry in the history curricula and textbooks for middle and high school students over three generations between 1955 and 2003. We argue that educational text functions as a litmus test to determine the importance of a subject in the national narrative. The article discusses two issues that reflect dilemmas in public, academic, and educational discourse: the representation of American Jews and their characteristics; and the representation of American religious movements and their principles. Two conclusions might be drawn from these: Firstly, those study topics have not been altered in the course of three generations, although the presentation of subject matter has been modified over the years; and secondly, there has been a significant change in the curricula and textbooks with regard to the Reform Movement. While the authors of the first-generation texts deny the legitimacy of the Reform Movement, those of the second-generation refer to it in neutral terms, thus conveying the impression that it is a legitimate phenomenon.

Empowering Lower Limb Disorder Identification through PoseNet and Artificial Intelligence.

A novel approach is presented in this study for the classification of lower limb disorders, with a specific emphasis on the knee, hip, and ankle. The research employs gait analysis and the extraction of PoseNet features from video data in order to effectively identify and categorize these disorders. The PoseNet algorithm facilitates the extraction of key body joint movements and positions from videos in a non-invasive and user-friendly manner, thereby offering a comprehensive representation of lower limb movements. The features that are extracted are subsequently standardized and employed as inputs for a range of machine learning algorithms, such as Random Forest, Extra Tree Classifier, Multilayer Perceptron, Artificial Neural Networks, and Convolutional Neural Networks. The models undergo training and testing processes using a dataset consisting of 174 real patients and normal individuals collected at the Tehsil Headquarter Hospital Sadiq Abad. The evaluation of their performance is conducted through the utilization of K-fold cross-validation. The findings exhibit a notable level of accuracy and precision in the classification of various lower limb disorders. Notably, the Artificial Neural Networks model achieves the highest accuracy rate of 98.84%. The proposed methodology exhibits potential in enhancing the diagnosis and treatment planning of lower limb disorders. It presents a non-invasive and efficient method of analyzing gait patterns and identifying particular conditions.

“Posthuman Intersections in BrexLit: Representing Migration in Contemporary British Fiction”

ABSTRACT The United Kingdom European Union membership or, in common parlance, Brexit referendum, held in the United Kingdom on 23rd June 2016, was an unprecedented event in European politics. This article analyses a literary corpus of long fiction, known as BrexLit, which have all drawn inspiration from this poll and its subsequent consequences for British society. Novels such as Ali Smith’s Autumn (2017), Anthony Cartwright’s The Cut (2017), Mohsin Hamid’s Exit West (2017), Jonathan Coe’s Middle England (2018), John Lanchester’s The Wall (2019) or Ian McEwan’s The Cockroach (2019), and others, will be examined through the lens of Posthuman Studies to establish the numerous intersections that can be identified within these texts’ realist and speculative worldviews. BrexLit questions traditional norms, such as the Humanist conception of the white man being the center of the universe, as well as it provides representation for migratory movements. In fact, migratory and refuge issues will form the centerpiece of this study as the majority of these novels challenge the abundant political discourses against the “Other” that have flourished before, during and after the Brexit campaign.

Nodes2STRNet for structural dense displacement recognition by deformable mesh model and motion representation

AbstractDisplacement is a critical indicator for mechanical systems and civil structures. Conventional vision‐based displacement recognition methods mainly focus on the sparse identification of limited measurement points, and the motion representation of an entire structure is very challenging. This study proposes a novel Nodes2STRNet for structural dense displacement recognition using a handful of structural control nodes based on a deformable structural three‐dimensional mesh model, which consists of control node estimation subnetwork (NodesEstimate) and pose parameter recognition subnetwork (Nodes2PoseNet). NodesEstimate calculates the dense optical flow field based on FlowNet 2.0 and generates structural control node coordinates. Nodes2PoseNet uses structural control node coordinates as input and regresses structural pose parameters by a multilayer perceptron. A self‐supervised learning strategy is designed with a mean square error loss and L2 regularization to train Nodes2PoseNet. The effectiveness and accuracy of dense displacement recognition and robustness to light condition variations are validated by seismic shaking table tests of a four‐story‐building model. Comparative studies with image‐segmentation‐based Structure‐PoseNet show that the proposed Nodes2STRNet can achieve higher accuracy and better robustness against light condition variations. In addition, NodesEstimate does not require retraining when faced with new scenarios, and Nodes2PoseNet has high self‐supervised training efficiency with only a few control nodes instead of fully supervised pixel‐level segmentation.

Stories in the Mind? The Role of Story-Based Categorizations in Motion Classification.

Categorization is fundamental for spatial and motion representation in both the domain of artificial intelligence and human cognition. In this paper, we investigated whether motion categorizations designed in artificial intelligence can inform human cognition. More concretely, we investigated if such categorizations (also known as qualitative representations) can inform the psychological understanding of human perception and memory of motion scenes. To this end, we took two motion categorizations in artificial intelligence, Motion-RCC and Motion-OPRA1 , and conducted four experiments on human perception and memory. Participants viewed simple motion scenes and judged the similarity of transformed scenes with this reference scene. Those transformed scenes differed in none, one, or both Motion-RCC and Motion-OPRA1 categories. Importantly, we applied an equal absolute metric change to those transformed scenes, so that differences in the similarity judgments should be due only to differing categories. In Experiments 1a and 1b, where the reference stimulus and transformed stimuli were visible at the same time (perception), both Motion-OPRA1 and Motion-RCC influenced the similarity judgments, with a stronger influence of Motion-OPRA1 . In Experiments 2a and 2b, where participants first memorized the reference stimulus and viewed the transformed stimuli after a short blank (memory), only Motion-OPRA1 had marked influences on the similarity judgments. Our findings demonstrate a link between human cognition and these motion categorizations developed in artificial intelligence. We argue for a continued and close multidisciplinary approach to investigating the representation of motion scenes.

Comparing Field, Probabilistic, and 2D Numerical Approaches to Assess Gravel Mobility in a Gravel‐Bed River

AbstractSediment transport is a key process that affects the morphology and ecological habitat diversity of rivers. As part of a gravel augmentation program to mitigate sediment deficit below a dam, gravel mobility in the Ain River in Eastern France was investigated by tracking of a large amount (n = 1,063) of PIT‐tagged gravels in the field, conducting a probabilistic approach based on published tracer studies, and performing two‐dimensional (2D) numerical modeling of flow and bedload transport. This comparative study highlights the strengths, weaknesses, and complementary aspects of the three approaches to the understanding of river gravel mobility. Thanks to recent technological improvements, PIT‐tagged gravels provide an increasingly reliable and accurate representation of bedload movement in the field, although it remains limited in spatio‐temporal resolution. Based on an exponential distribution, the probabilistic approach correctly reproduces the average trend in travel distances by the different classes of particles over hydrological periods, including one or several significant floods. Furthermore, the 2D numerical modeling accounts for the variability of local hydrodynamic conditions and can simulate realistic displacement distributions for the different classes of particles with high spatio‐temporal resolution. Numerical modeling is a very encouraging approach, which makes our study original because it is the first time that the estimation of mean travel distances, the application of an exponential distribution, and the comparison with a hydrodynamic model are combined. A more effective modeling strategy involves incorporating a probabilistic transport model in the 2D numerical model to reproduce the observed scatter of the individual particle trajectories.

Spatiotemporal Consistency Learning From Momentum Cues for Human Motion Prediction

Extrapolating future human motion based on the historical human pose sequence is the foundation of various intelligent applications. Numerous deep learning-based algorithms have been designed to address this task, achieving state-of-the-art performance on different human motion benchmark datasets. However, most existing methods employ three-dimensional coordinates of joints to demonstrate dynamic motion contexts implicitly. Unfortunately, it remains challenging in capturing motion information from the pose sequence. In this paper, we advocate explicitly describing dynamic contexts via the momentum of human motion mechanic space, as the momentum of a joint is explicit, temporal consistent, and can provide abundant information to the model. In addition, the single-stream methods play a dominant role in the field of human motion prediction. They usually capture motion information via the strategy of continuous or sparse sampling, which might obviate global or detailed local information. Therefore, we present a simple yet effective dual-stream method that can consider both the detailed and global temporal information through a combination of continuous and sparse sampling. The proposed dual-stream paradigm enables the improvement of computational efficiency and the short-term prediction accuracy concurrently. Furthermore, we present a novel temporal attention-based graph convolutional network (TA-GCN) to derive a spatiotemporally consistent motion representation, which can adequately consider the rationality of human body topology. Extensive experiments on two large motion prediction benchmark datasets ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e</i> ., Human 3.6M and CMU Mocap) show that our algorithm achieves state-of-the-art performance both qualitatively and quantitatively.

Effects of Motion-Relevant Knowledge From Unlabeled Video to Human-Object Interaction Detection.

The existing works on human-object interaction (HOI) detection usually rely on expensive large-scale labeled image datasets. However, in real scenes, labeled data may be insufficient, and some rare HOI categories have few samples. This poses great challenges for deep-learning-based HOI detection models. Existing works tackle it by introducing compositional learning or word embedding but still need large-scale labeled data or extremely rely on the well-learned knowledge. In contrast, the freely available unlabeled videos contain rich motion-relevant information that can help infer rare HOIs. In this article, we creatively propose a multitask learning (MTL) perspective to assist in HOI detection with the aid of motion-relevant knowledge learning on unlabeled videos. Specifically, we design the appearance reconstruction loss (ARL) and sequential motion mining module in a self-supervised manner to learn more generalizable motion representations for promoting the detection of rare HOIs. Moreover, to better transfer motion-related knowledge from unlabeled videos to HOI images, a domain discriminator is introduced to decrease the domain gap between two domains. Extensive experiments on the HICO-DET dataset with rare categories and the V-COCO dataset with minimum supervision demonstrate the effectiveness of motion-aware knowledge implied in unlabeled videos for HOI detection.

The Vanishing Coin Illusion: When sound congruence affects visual representation of motion.

In the "classic" vanishing ball illusion (VBI), a magician pretends to throw a ball and the audience sees it go up and then disappear. To gain a better understanding of the mechanisms involved, we conducted two experiments to analyze the permeability of this type of illusion to auditory information. A modified version of the VBI (i.e., the Vanishing Coin Illusion [VCI]), was presented, and this was either accompanied by a sound or was not. The results show that the presence of a sound adds to the success of the illusion (Experiment 1), especially when this sound is congruent with the illusion (Experiment 2). Based on these results, we discuss the mechanisms at work in this illusion.

Bodily expressed emotion understanding through integrating Laban movement analysis

Bodily expressed emotion understanding (BEEU) aims to automatically recognize human emotional expressions from body movements. Psychological research has demonstrated that people often move using specific motor elements to convey emotions. This work takes three steps to integrate human motor elements to study BEEU. First, we introduce BoME (body motor elements), a highly precise dataset for human motor elements. Second, we apply baseline models to estimate these elements on BoME, showing that deep learning methods are capable of learning effective representations of human movement. Finally, we propose a dual-source solution to enhance the BEEU model with the BoME dataset, which trains with both motor element and emotion labels and simultaneously produces predictions for both. Through experiments on the BoLD in-the-wild emotion understanding benchmark, we showcase the significant benefit of our approach. These results may inspire further research utilizing human motor elements for emotion understanding and mental health analysis.

Dogs Rely On Visual Cues Rather Than On Effector-Specific Movement Representations to Predict Human Action Targets.

The ability to predict others' actions is one of the main pillars of social cognition. We investigated the processes underlying this ability by pitting motor representations of the observed movements against visual familiarity. In two pre-registered eye-tracking experiments, we measured the gaze arrival times of 16 dogs (Canis familiaris) who observed videos of a human or a conspecific executing the same goal-directed actions. On the first trial, when the human agent performed human-typical movements outside dogs' specific motor repertoire, dogs' gaze arrived at the target object anticipatorily (i.e., before the human touched the target object). When the agent was a conspecific, dogs' gaze arrived to the target object reactively (i.e., upon or after touch). When the human agent performed unusual movements more closely related to the dogs' motor possibilities (e.g., crawling instead of walking), dogs' gaze arrival times were intermediate between the other two conditions. In a replication experiment, with slightly different stimuli, dogs' looks to the target object were neither significantly predictive nor reactive, irrespective of the agent. However, when including looks at the target object that were not preceded by looks to the agents, on average dogs looked anticipatorily and sooner at the human agent's action target than at the conspecific's. Looking times and pupil size analyses suggest that the dogs' attention was captured more by the dog agent. These results suggest that visual familiarity with the observed action and saliency of the agent had a stronger influence on the dogs' looking behaviour than effector-specific movement representations in anticipating action targets.

Keypoint Changes for Fast Human Activity Recognition

Human activity recognition has been an open problem in computer vision for almost 2 decades. During this time, there have been many approaches proposed to solve this problem, but very few have managed to solve it in a way that is sufficiently computationally efficient for real-time applications. Recently, this has changed, with keypoint-based methods demonstrating a high degree of accuracy with low computational cost. These approaches take a given image and return a set of joint locations for each individual within an image. In order to achieve real-time performance, a sparse representation of these features over a given time frame is required for classification. Previous methods have achieved this using a reduced number of keypoints, but this approach gives a less robust representation of the individual’s body pose and may limit the types of activity that can be detected. We present a novel method for reducing the size of the feature set, by calculating the Euclidian distance and the direction of keypoint changes across a number of frames. This allows for a meaningful representation of the individuals movements over time. We show that this method achieves accuracy on par with current state-of-the-art methods, while demonstrating real-time performance.

Dissociating representations of affect and motion in visual cortices.

While a delicious dessert being presented to us may elicit strong feelings of happiness and excitement, the same treat falling slowly away can lead to sadness and disappointment. Our emotional response to the item depends on its visual motion direction. Despite this importance, it remains unclear whether (and how) cortical areas devoted to decoding motion direction represents or integrates emotion with perceived motion direction. Motion-selective visual area V5/MT+ sits, both functionally and anatomically, at the nexus of dorsal and ventral visual streams. These pathways, however, differ in how they are modulated by emotional cues. The current study was designed to disentangle how emotion and motion perception interact, as well as use emotion-dependent modulation of visual cortices to understand the relation of V5/MT+ to canonical processing streams. During functional magnetic resonance imaging (fMRI), approaching, receding, or static motion after-effects (MAEs) were induced on stationary positive, negative, and neutral stimuli. An independent localizer scan was conducted to identify the visual-motion area V5/MT+. Through univariate and multivariate analyses, we demonstrated that emotion representations in V5/MT+ share a more similar response profile to that observed in ventral visual than dorsal, visual structures. Specifically, V5/MT+ and ventral structures were sensitive to the emotional content of visual stimuli, whereas dorsal visual structures were not. Overall, this work highlights the critical role of V5/MT+ in the representation and processing of visually acquired emotional content. It further suggests a role for this region in utilizing affectively salient visual information to augment motion perception of biologically relevant stimuli.

Representation of Southern African National Liberation Movements at the United Nations, 1972-1974

Representation of Southern African National Liberation Movements at the United Nations, 1972-1974

Air traffic controllers' mental fatigue recognition: A multi-sensor information fusion-based deep learning approach

With the growing density of air passenger traffic, accurately recognizing the level of mental fatigue (MF) experienced by air traffic controllers (ATCOs) is crucial for developing intelligent ATCOs' mental state monitoring systems, which can achieve a more effective and safer human–machine cooperative pattern. However, the existing methods for recognizing ATCOs' MF face significant challenges due to pattern variations between ATCOs and sensor artifacts. This study introduces a framework for ATCOs' MF recognition, utilizing a deep neural network called RecMF, which incorporates multi-sensor information fusion to enhance the performance of MF detection. Specifically, the RecMF employs an attention-enabled CNN-LSTM architecture that simultaneously captures time-series feature representations of electroencephalogram (EEG) signals and eye movements. To validate the effectiveness of RecMF, a fatigue-inducing experiment is conducted involving 28 subjects who are tasked with performing a series of air traffic control (ATC) tasks. The model's performance is evaluated across various time horizons and typical cognitive tasks to gain insights into its capabilities. The evaluation results indicate that the proposed model outperforms other existing methods, thereby confirming its feasibility and effectiveness. Additionally, the effects of MF on ATCOs' cognitive performance are analyzed using analysis of variance (ANOVA). The results reveal that higher levels of MF significantly reduce ATCOs' reaction speed and operational accuracy.

Review for Optimal Human-Gesture Design Methodology and Motion representation of Medical Images using segmentation from depth data and gesture recognition.

Human gesture recognition and motion representation has become a vital base of current intelligent human-machine interfaces because of ubiquitous and more comfortable interaction. Human-Gesture recognition chiefly deals with recognizing meaningful, expressive body movements involving physical motions of face, head, arms, fingers, hands or body. This review article presents a concise overview of optimal human-gesture and motion representation of medical images. This paper surveys various works undertaken on human gesture design and discusses various design methodologies used for image segmentation and gesture recognition. It further provides a general idea of modeling techniques for analysing hand gesture images and even discusses the diverse techniques involved in motion recognition. This survey provides an insight into various efforts and developments made in the gesture/motion recognition domain through analyzing and reviewing the procedures, datasets, recognition rates and approaches employed for identifying diverse human motions and gestures for supporting better and devising improved applications in near future.

Enriched Rehabilitation Reduces Abnormal Motor Synergies and Enhances Motor Plasticity Following Severe Stroke in Rats.

Stroke results in loss of upper motor neuron control over voluntary movements and emergence of abnormal synergies. Presently, it is unclear to what extent poststroke recovery reflects true recovery (restitution), compensation, or some combination of these processes. Here, we investigated this question using behavioral and kinematic analyses of skilled reaching in rats subjected to severe stroke that affected both the forelimb motor cortex and dorsolateral striatum. After stroke, male rats either spontaneously recovered or received enriched rehabilitation. We assessed forelimb motor recovery using behavioral and kinematic outcome measures. To provide insights into the mechanisms underlying the effects of rehabilitation on behavior, we used intracortical microstimulation and FosB (protein fosB) immunostaining techniques. Enriched rehabilitation significantly improved food pellet retrieval in the staircase-reaching task. Rehabilitation resulted in several poststroke flexion synergies returning to prestroke patterns, and across subjects, these changes correlated with the intensity of rehabilitation. Enriched rehabilitation increased the proportion of distal movement representation in the perilesional cortex and increased use-dependent activation in the ipsilesional red nucleus. These results provide evidence that enriched rehabilitation enhances recovery, at least in part, by restitution of forelimb function following severe stroke. Furthermore, the restitution of function is associated with changes in multiple motor-related structures at different levels of the central nervous system. A better understanding of the processes that underlie improved motor performance, along with the identification of midbrain circuits activated by rehabilitation, represent new insights and potential targets for optimizing poststroke recovery.