Static Frame Research Articles

Consistency-constrained unsupervised video anomaly detection framework based on Co-teaching

Intelligent video surveillance continues to be a vibrant research domain within the field of computer vision. However, existing representation learning frameworks primarily focus on static information extraction frame by frame such as appearance features, they often overlook the valuable dynamic information like optical flow feature inherent in the video data, which is most essential characteristics of sequence data. To mining dynamic features and bridge this gap, our paper introduces a novel anomaly detection framework that balance dynamic information with static information and construct a relationship between appearance features and corresponding optical flow features, where we sets strong consistency constraints, which reduce the loss between dynamic information and corresponding static information, and we leverages collaborative teaching network to ensure a consistent representation of both static and dynamic information for predict. The proposed framework consists of two sets of encoder–decoder pairs complemented by a memory storage module. Operating in parallel with the dual encoder network is a Co-teaching network, with the shared memory module serving as the cornerstone for collaborative training. The Consistency constrained condition guarantees the strong consistency of dynamic and static information in the learned representations. In our experimental phase, we present compelling results that showcase the superior performance of our algorithm across three publicly available datasets.

Fluctuating wind load encountered by linearly moving vehicles: Investigation on fluctuating wind spectral characteristics

Wind load is one of the controlling loads for moving vehicles. Existing studies on the wind-induced response of vehicles mostly apply traditional wind spectra directly to the vehicles, overlooking their motion status. Fluctuating wind characteristics encountered by linearly moving vehicles were investigated in this paper. The traditional static frame of view was adjusted to a moving one, two principal flow directions were considered and the idea of positional equivalence introduced in Taylor frozen assumption was extended to the moving direction. A fluctuating wind power spectral density (PSD) was proposed. Then a novel test was designed to enable a measuring probe to travel across a wind field. The fluctuating wind characteristics encountered by moving points are quite different from those encountered by a static point. The integral scale decreases as the speed ratio of the measuring point to the wind increases, and an elliptic model was suggested. The proposed PSD aligns well with the test results, indicating that it can be used to accurately characterize the fluctuating wind load encountered by moving vehicles. Both the test results and the numerical analysis showed that as the speed ratio increases, the PSD and energy spectrum as a whole migrate to higher frequencies.

Event-Oriented State Alignment Network for Weakly Supervised Temporal Language Grounding.

Weakly supervised temporal language grounding (TLG) aims to locate events in untrimmed videos based on natural language queries without temporal annotations, necessitating a deep understanding of semantic context across both video and text modalities. Existing methods often focus on simple correlations between query phrases and isolated video segments, neglecting the event-oriented semantic coherence and consistency required for accurate temporal grounding. This can lead to misleading results due to partial frame correlations. To address these limitations, we propose the Event-oriented State Alignment Network (ESAN), which constructs "start-event-end" semantic state sets for both textual and video data. ESAN employs relative entropy for cross-modal alignment through knowledge distillation from pre-trained large models, thereby enhancing semantic coherence within each modality and ensuring consistency across modalities. Our approach leverages vision-language models to extract static frame semantics and large language models to capture dynamic semantic changes, facilitating a more comprehensive understanding of events. Experiments conducted on two benchmark datasets demonstrate that ESAN significantly outperforms existing methods. By reducing false high correlations and improving the overall performance, our method effectively addresses the challenges posed by previous approaches. These advancements highlight the potential of ESAN to improve the precision and reliability of temporal language grounding tasks.

Interaction of stream-wise vortices generated by swirler grid

A system of stream-wise vortices has been created using a grid of swirling elements with alternating orientations (like a chessboard). The particle image velocimetry method has been used to map the velocity field in several planes perpendicular to the stream. The mesh-based Reynolds number is 1.35×104 and 2.71×104, respectively. The stream-wise development of turbulent kinetic energy (TKE) shows first an increase in a distance of x≈10M, followed by power-law decay. Individual vortices are detected in each snapshot. The radial profile of TKE transformed to a vortex coordinate system is almost constant, either with maximum as in static frame or zero as observed by previous research. The properties of detected vortices are studied statistically: the meandering amplitude expressed as the standard deviation of vortex positions grows roughly as ∼ex, i.e., faster than expected random-walk growth ∼x. Vortex circulation decays exponentially as predicted by classical Helmholtz theorem. The interaction between neighboring vortices is expressed via correlation of selected quantities. Correlation of energy develops downstream from anticorrelation to a positive correlation. The strongest correlation is observed between the first vortex circulation and the second vortex position perpendicular to their connection line. Other correlations are weak.

Semantics and structure of the concept of “Bogatyrstvo” in epics and modern linguistic consciousness

The subject of the study is the concept of "Bogatyrstvo" and its structural and semantic features. Bogatyrstvo is inextricably linked with the hero and with the plot of epics. In epics, it is perceived as a concept-script, that is, it is formed in the process of the entire narrative. However, the original structure of the concept is presented as a conceptual field of meanings inscribed in a more complex structural formation – the ethical concept of "Courage". The relevance of the research is dictated by the high precedent of the name of the concept and its dominant verbalizers. The semantic field of the concept is explored in etymological, historical and topical layers. The main semantic load of the etymological layer of the concept is formed in the Church Slavonic language. The historical layer is vividly represented in the dictionaries of the XIX and XX centuries, the latter captures the limitation of the "existence" of the concept by the genre of epic. The actual layer is revealed in the process of studying the consciousness of modern students. The purpose of the article is to identify the reference points (slots) that formed the basis of the concept. A special contribution of the authors and the novelty of the study is the conduct of a semantic experiment, which established that in the minds of modern native speakers, the concept Bogatyrstvo, for which analysis methods were used (analysis of dictionary definitions and contextual analysis of epics), semantic analysis (the "supporting" semantic shares of the concept of "Bogatyrstvo" were identified, their key verbalizers were highlighted) and the method of semantic experiment to identify the modern perception of the concept. The main conclusions of the study: 1) the concept of "Bogatyrstvo" has a static frame structure and a field structure, including slots "spiritual strength" and "physical strength". Spiritual strength lies in the constant readiness to stand up for the motherland as a manifestation of true courage and heroism; in the fate of the hero-defender, ordained from above; in the Christian faith. The physical strength of the hero is, as it were, the personification of the whole strength of the Russian people, given by God. 2) These slots are present in the epics in the dynamic structure of the script: the slot "physical force" unfolds to the whole epic action; the slot "spiritual force" is expressed through the entire epic narrative – this is the very motivation for action. A special contribution of the authors and the novelty of the study is the conduct of a semantic experiment, which established that in the minds of modern native speakers, the concept of "Bogatyrstvo" is presented statically and includes both slots, with the predominance of the slot "spiritual power".

The Theory of Absoluteness —The Relations among Matter, Space, Time and Motion

In this paper, the absolute static reference frame of the universe is established, the Galileo's relativity principle and the essence of light are analyzed, and the zero result of Michelson-Morey experiment is re-recognized. The conclusion is that the light in two directions reaches the eyepiece at the same time, and when the interferometer is rotated at any angle, the light in two directions still reaches the eyepiece at the same time, and the interference fringes will not move, so the zero result of the experiment is inevitable. This paper analyzes Lorenz's explanation of the zero result of the Michelson-Morey experiment, and points out that there are many contradictions and mistakes in it. Michelson confused the reference frame in his analysis and calculation, and the result was necessarily wrong. The Michelson-Morey experiment wants to measure the speed of the earth relative to the "ether" on the earth. According to the Galileo's relativity principle, it is impossible to measure it. This experiment is meaningless. Because Michelson's wrong calculation result is not consistent with the zero result of the experiment, Lorenz had to set up "length shortening", "clock slowing down" and a Lorenz transformation factor to piece together the zero result of the experiment.It is precisely because these assumptions are incorrect that all the relations and related theories that use Lorentz transformation factors are incorrect, and the special theory of relativity is completely wrong. Only the Galileo's relativity principle can correctly explain the zero result of Michelson-Morey experiment, and the Galileo’s transformation is the correct theory to describes motion. This paper also re-understands the relations among matter, space, time, motion, inertia, field, field wave, force and energy, analyzes the essence of nuclear energy, and establishes a new model of the universe. It makes physics return to the correct track of classical theory and deepens and develops classical theory.

The Theory of Absoluteness —The Relations among Matter, Space, Time and Motion

In this paper, the absolute static reference frame of the universe is established, the Galileo's relativity principle and the essence of light are analyzed, and the zero result of Michelson-Morey experiment is re-recognized. The conclusion is that the light in two directions reaches the eyepiece at the same time, and when the interferometer is rotated at any angle, the light in two directions still reaches the eyepiece at the same time, and the interference fringes will not move, so the zero result of the experiment is inevitable. This paper analyzes Lorenz's explanation of the zero result of the Michelson-Morey experiment, and points out that there are many contradictions and mistakes in it. Michelson confused the reference frame in his analysis and calculation, and the result was necessarily wrong. The Michelson-Morey experiment wants to measure the speed of the earth relative to the "ether" on the earth. According to the Galileo's relativity principle, it is impossible to measure it. This experiment is meaningless. Because Michelson's wrong calculation result is not consistent with the zero result of the experiment, Lorenz had to set up "length shortening", "clock slowing down" and a Lorenz transformation factor to piece together the zero result of the experiment.It is precisely because these assumptions are incorrect that all the relations and related theories that use Lorentz transformation factors are incorrect, and the special theory of relativity is completely wrong. Only the Galileo's relativity principle can correctly explain the zero result of Michelson-Morey experiment, and the Galileo’s transformation is the correct theory to describes motion. This paper also re-understands the relations among matter, space, time, motion, inertia, field, field wave, force and energy, analyzes the essence of nuclear energy, and establishes a new model of the universe. It makes physics return to the correct track of classical theory and deepens and develops classical theory

Кинематограф в одном кадре

The article examines the status of the screenshot, a separate static frame, a fragment of the cinematic stream, its relationship to the “whole” of the film, its influence on the spectator’s perception, as well as everyday user practices involving the observation and creation of screenshots. The starting point of the text is an attempt to reconsider the common notion of the screenshot/one-frame as a particle in a subordinate position to the film (and more broadly to the history of cinema). On the basis of texts by Pascal Bonitzer, Sergei Eisenstein, and Roland Barthes, the basic thesis is formulated — the mutual relation between the frame and the film can be constructed differently than the logic of “the frame is a lower-order element nested within the whole film;” on the contrary, the frame and the film in some cases meet as equivalent phenomena or they are superimposed on each other according to the principle of palimpsest. The article deals with three common forms of using frames and screenshots advertising and illustrative shots, the use of film images in sticker packs for messengers, and albums of images in social networks. The question is raised about the communicative plasticity of separate images, the metamorphoses that occur with them when they are taken out of their original context, and their insertion into new artistic and research practices. As such practices, the text pays special attention to two cases — the screenshot project of French director Frank Beauvais and contemporary audiovisual film criticism (based on the works of Johannes Binotto, Carrie Griffith, Daniel McIlwright).

TubULAR: tracking in toto deformations of dynamic tissues via constrained maps.

A common motif in biology is the arrangement of cells into tubes, which further transform into complex shapes. Traditionally, analysis of dynamic tissues has relied on inspecting static snapshots, live imaging of cross-sections or tracking isolated cells in three dimensions. However, capturing the interplay between in-plane and out-of-plane behaviors requires following the full surface as it deforms and integrating cell-scale motions into collective, tissue-scale deformations. Here, we present an analysis framework that builds in toto maps of tissue deformations by following tissue parcels in a static material frame of reference. Our approach then relates in-plane and out-of-plane behaviors and decomposes complex deformation maps into elementary contributions. The tube-like surface Lagrangian analysis resource (TubULAR) provides an open-source implementation accessible either as a standalone toolkit or as an extension of the ImSAnE package used in the developmental biology community. We demonstrate our approach by analyzing shape change in the embryonic Drosophila midgut and beating zebrafish heart. The method naturally generalizes to in vitro and synthetic systems and provides ready access to the mechanical mechanisms relating genetic patterning to organ shape change.

Poster Session I: Dissociation of torsional eye movements and perception during optokinetic stimulation by visual cues of gravity.

An optokinetic stimulus rotating about the naso-occipital axis drives torsional eye movements and causes a bias in perception of upright measured with a subjective visual vertical (SVV) task. In addition, a static image with tilted visual cues of gravity orientation will induce optostatic torsion and biases SVV. We posit that a visual gravity cue provided by a frame combined with a torsional optokinetic stimulus would increase measured torsion and further bias SVV. We use a VR headset with eye-tracking to place the subject in a virtual room with circles forming either a rectangular room (frame condition) or a tubular room (no-frame condition). We place a fixation point at the center of the room while the subject performs a SVV task. The room either rotates about the naso-occipital axis at 0.05 Hz, 0.1 Hz, or 0.2 Hz, with an amplitude of ±20°, or has a static tilt of 0° or ±30°. Static frame conditions had the expected bias of SVV and optostatic torsion compared to the control no-frame condition. In sinusoidal rotation conditions, for SVV, there was a significant difference in the amplitude of the response between the frame (5.4 ± 2.6°; mean ± std) and no-frame (3.0 ± 1.3°) condition, while there was no significant difference for torsion, frame (0.8 ± 0.6°) and no-frame (0.7 ± 0.6°). Our results indicate that while perception integrates a moving visual cue into its estimation of upright orientation, the ocular motor system is only affected by those cues when they are static.

Correction and integration of solid-angle data from the azimuthally resolving 2D detector at ID06-LVP, ESRF.

The unique diffraction geometry of ESRF beamline ID06-LVP offers continuous static 2D or azimuthally resolving data collections over all accessible solid angles available to the tooling geometry. The system is built around a rotating custom-built Pilatus3 CdTe 900k-W detector from Dectris, in a configuration equivalent to three butted 300k devices. As a non-standard geometry, here the method of alignment, correction and subsequent integration for any data collected over all solid angles accessible, or over any azimuthal range contained therein, are provided and illustrated by parameterizing and extending existing pyFAI routines. At 1° integrated intervals, and typical distances (2.0 m), the system covers an area of near 2.5 m2 (100 Mpx square equivalent), to 0.65 Å resolution, at 53 keV from a total dataset of some 312 Mpx. Standard FWHMs of SRM660a LaB6 vary from 0.005° to 0.01°, depending on beam size, energy and sample dimensions, and are sampled at an elevated rate. The azimuthal range per static frame ranges from <20° to ∼1° over the full range of the detector surface. A full 2θ-intensity data collection at static azimuth takes 1-3 s typically, and can be reduced to ms-1 rates for measurements requiring time-rate determination. A full solid-angle collection can be completed in a minute. Sample detector distances are accessible from 1.6 m to 4.0 m.

Revealing the effect of phase and time coupling on NMR relaxation rate by random walks in phase space.

Phase-time coupling is a natural process in the phase random walks of a spin system; however, its effect on the nuclear magnetic resonance (NMR) relaxation is a challenge to the established theories such as the second-order quantum perturbation theory. This paper extends the recently developed phase diffusion method to treat the phase-time coupling effect, based on uncoupled phase diffusions, and coupled random walks. The instantaneous projection of the rotating random field is employed to get the accumulated phase of the NMR observable. In the static frame and the rotating frame, the phase diffusion coefficients are derived. The obtained theoretical results show that the phase-time coupling has a significant impact on the NMR relaxation rate: The angular frequency ω in the spectral density is modified to an apparent angular frequency ηω, where η is the phase-time coupling constant. The strongest coupling has η equaling 2, while η equaling 1 corresponds to the traditional results. As an example, the modified relaxation time expressions based on both monoexponential and nonmonoexponential functions can successfully fit the previously reported ^{13}CT_{1} NMR experimental data of polyisobutylene (PIB) in the blend of PIB and head-to-head poly(propylene). In contrast, the traditional relaxation rate expression based on the monoexponential time correlation function cannot fit such experimental data. With the phase-time coupling, the obtained characteristic time of the segmental motion is faster than that from conventional results.

Directional Residual Frame: Turns the motion information into a static RGB frame

Directional Residual Frame: Turns the motion information into a static RGB frame

Sensing Mechanism and Real-Time Bridge Displacement Monitoring for a Laboratory Truss Bridge Using Hybrid Data Fusion

Remote and real-time displacement measurements are crucial for a successful bridge health monitoring program. Researchers have attempted to monitor the deformation of bridges using remote sensing techniques such as an accelerometer when a static reference frame is not available. However, errors accumulate throughout the double-integration process, significantly reducing the reliability and accuracy of the displacement measurements. To obtain accurate reference-free bridge displacement measurements, this paper aims to develop a real-time computing algorithm based on hybrid sensor data fusion and implement the algorithm via smart sensing technology. By combining the accelerometer and strain gauge measurements in real time, the proposed algorithm can overcome the limitations of the existing methods (such as integration errors, sensor drifts, and environmental disturbances) and provide real-time pseud-static and dynamic displacement measurements of bridges under loads. A wireless sensor, SmartRock, containing multiple sensing units (i.e., triaxial accelerometer and strain gauges) and a Micro Controlling Unit (MCU) were utilized for remote data acquisition and signal processing. A remote sensing system (with SmartRocks, an antenna, an industrial computer, a Wi-Fi hotspot, etc.) was deployed, and a laboratory truss bridge experiment was conducted to demonstrate the implementation of the algorithm. The results show that the proposed algorithm can estimate a bridge displacement with sufficient accuracy, and the remote system is capable of the real-time monitoring of bridge deformations compared to using only one type of sensor. This research represents a significant advancement in the field of bridge displacement monitoring, offering a reliable and reference-free approach for remote and real-time measurements.

Two-Dimensional Flow on the Sphere

Equilibrium statistical mechanics predicts that inviscid, two-dimensional, incompressible flow on the sphere eventually reaches a state in which spherical harmonic modes of degrees n=1 and n=2 hold all the energy. By a separate theory, such flow is static in a reference frame rotating at angular speed 2Ω/3 with respect to the inertial frame. The vorticity field in the static frame is an accident of the initial conditions, but, once established, it lasts forever under the stated assumptions. We investigate the possibility of such behavior with a stereographic-coordinate model that conserves energy and enstrophy when the viscosity vanishes.

ПРИЛИВНЫЕ СИЛЫ ВБЛИЗИ ЧЕРНЫХ ДЫР В ГРАВИТАЦИОННОЙ ТЕОРИИ С НАРУШЕНИЕМ СИММЕТРИИ ЛОРЕНЦА

It is generally accepted that the curvature of space-time is small near the event horizon of a large static black hole, and test particles can fall into it without being destroyed. However, in most cases, at a small distance from the event horizon, the curvature of the black hole can be so huge that the test particle will stretch and even collapse under the influence of tidal forces. Thus, any object that falls into black holes experiences huge tidal forces beyond the horizon. This is a result of the fact that the curvature is very large near black holes and almost zero near the horizon. When measurements are made in a static frame (which also becomes zero), the curvature components remain small. This means that all curvature invariants are small, and any perturbing corrections, such as charge, angular momentum, Ricci scalar, etc., can be neglected. However, in a freely falling frame (the reference frame of a distant asymptotic observer), the curvature components are very large. In our paper, we consider the influence of the perturbing parameter ℓ on the curvature invariants of black holes in the gravitational theory with Lorentz symmetry breaking. The results are compared with the Schwarzschild black hole solution, which can be obtained by substituting ℓ =0.

Deep learning based registration for head motion correction in positron emission tomography as a strategy for improved image quantification

Objectives: Positron emission tomography (PET) is affected by various kinds of patient movement during a scan. Frame-by-frame image registration is one of the most practiced motion correction techniques. In recent years, deep learning has shown a remarkable ability to quickly and accurately register images once trained. This paper studies the feasibility of using a deep learning framework to correct 3D positron emission tomography image volumes for head motion in routine positron emission tomography imaging to improve quantification in motion impacted data.Materials and Methods: A neural network was trained with 3D positron emission tomography image volumes in an unsupervised manner to predict transformation parameters required to perform image registration. A multi-step convolutional neural network (CNN) was combined with a spatial transform layer. Pairs of target and source images were used as input to the network. To prepare the training dataset, a previously published TOF-PEPT algorithm was applied to automatically detect static frames where the patient remained in a relatively steady position and transitional frames where they underwent abrupt motion. A single image volume was reconstructed for each static frame. The image reconstructed from the first static frame served as the target image with images from subsequent static frames being used as source images. The trained neural network predicted transformation parameters that could be used to perform frame-by-frame image-based motion correction but also enabled raw listmode positron emission tomography data correction where individual line-of-responses were repositioned. Line profiles and ROIs were drawn across the reconstructed image volumes to compare performance and quantitative results between standard registration tools and the deep learning technique. Corrected volumes were further compared to motion free images quantitatively using Dice indices.Results: In total, one hundred 3D positron emission tomography image volumes were used to train the network. Cross-validation was carried out using a 4:1 ratio for the training and test data. A conventional algorithm for affine registration from the Advanced Normalization Tools (ANTs) software package served as a baseline. To evaluate the correction performance, the mean Dice index and standardized uptake value (SUV) were used. Application of the algorithm to clinical data showed good performance with respect to registration accuracy as well as processing time. The neural network yielded a mean Dice index of ∼0.87 which was similar to the advanced Normalization Tools algorithm and did so ∼3x faster using a multi-core CPU and ∼20x faster with a GPU. Standardized uptake value analysis showed that quantitative results were 30%–60% higher in the motion-corrected images, and the neural network performed better than or close to the advanced Normalization Tools.Conclusion: The aim of this work was to study the quantitative impact of using a data-driven deep learning motion correction technique for positron emission tomography data and assess its performance. The results showed the technique is capable of producing high quality registrations that compensate for patient motion that occurs during a scan and improve quantitative accuracy.

Starting Point Selection and Multiple-Standard Matching for Video Object Segmentation With Language Annotation

In this paper, we study the language-level video object segmentation where the first-frame language annotation is provided to describe the target object. By taking full advantage of the characteristic that a language label is normally compatible to all frames in a video, the proposed method can choose the most suitable starting frame to mitigate the initialization failure issue. Moreover, apart from extracting the visual feature from a static video frame, a motion-language score based on optical flow is proposed to better represent moving objects. Ultimately, scores of multiple standards are aggregated using an attention-based mechanism to predict the final result. The proposed method is evaluated on four widely-used video object segmentation datasets, including DAVIS 2017, DAVIS 2016, SegTrack V2 and YoutubeObject datasets, and the new state-of-the-art accuracy (mean region similarity) is obtained on both DAVIS 2017 (67.2%) and DAVIS 2016 (83.5%) datasets. Source code will be published together with the paper.

A Novel Gripper with Integrated Rotary Unit and Force Control for Pick and Place Applications

Modern electrical grippers have lower life-cycle costs compared to pneumatic ones. Furthermore, they provide force control, making it possible to grasp objects with different fragility using a single device. At the same time, electrical grippers have a higher end-effector weight, installed on the robot’s flange and lower closing speed, preventing them from replacing pneumatic solutions in high dynamic Pick and Place applications. This research faces both issues by synthesizing a novel gripper mechanism based on a Torque Distribution Gearbox, which makes it possible to relocate the electric motors to the static frame of a delta robot. The proposed gripper not only has a lower mass and a higher closing speed than competitive electric solutions, but it also provides unlimited rotation around the vertical axis. The performance of the gripper was tested in experimental studies, which showed that a created aluminum prototype provides a precise force control in the range from 3 N to 48 N with an accuracy not worse than 1.27 N. Moreover, its finger’s speed is 3.1–56 times higher than market available electrical grippers, which makes it comparable by this parameter with pneumatic solutions used in high dynamic Pick and Place applications.

Research on resonance mechanism and damping method of grid-connected inverter with LCL filter for battery energy storage system

To improve the stability of the grid-connected of the battery energy storage system, Firstly, a mathematical model of the inverter with current feedback control on the inverter side is established in a two-phase static frame. Secondly, based on the inverter model, the resonance mechanism of the inverter grid-connected system is studied. By means of the combination of the digital control and passive damping, a novel method of inverter output admittance reshaping is proposed afterward. The procedure of the novel method is given as well as the exemplary applications. The validity of the inverter-grid resonance mechanism and the effectiveness of the novel method to damp the resonance is identified by simulation results.