Object Motion Research Articles

MotionTrack: Learning motion predictor for multiple object tracking

Significant progress has been achieved in multi-object tracking (MOT) through the evolution of detection and re-identification (ReID) techniques. Despite these advancements, accurately tracking objects in scenarios with homogeneous appearance and heterogeneous motion remains a challenge. This challenge arises from two main factors: the insufficient discriminability of ReID features and the predominant utilization of linear motion models in MOT. In this context, we introduce a novel motion-based tracker, MotionTrack, centered around a learnable motion predictor that relies solely on object trajectory information. This predictor comprehensively integrates two levels of granularity in motion features to enhance the modeling of temporal dynamics and facilitate precise future motion prediction for individual objects. Specifically, the proposed approach adopts a self-attention mechanism to capture token-level information and a Dynamic MLP layer to model channel-level features. MotionTrack is a simple, online tracking approach. Our experimental results demonstrate that MotionTrack yields state-of-the-art performance on datasets such as Dancetrack and SportsMOT, characterized by highly complex object motion.

Head tracking using an optical soft tactile sensing surface.

This research proposes a sensor for tracking the motion of a human head via optical tactile sensing. It implements the use of a fibrescope a non-metal alternative to a webcam. Previous works have included robotics grippers to mimic the sensory features of human skin, that used monochrome cameras and depth cameras. Tactile sensing has shown advantages in feedback-based interactions between robots and their environment. The methodology in this paper is utilised to track motion of objects in physical contact with these sensors to replace external camera based motion capture systems. Our immediate application is related to detection of human head motion during radiotherapy procedures. The motion was analysed in two degrees of freedom, respective to the tactile sensor (translational in z-axis, and rotational around y-axis), to produce repeatable and accurate results. The movements were stimulated by a robot arm, which also provided ground truth values from its end-effector. The fibrescope was implemented to ensure the device's compatibility with electromagnetic waves. The cameras and the ground truth values were time synchronised using robotics operating systems tools. Image processing methods were compared between grayscale and binary image sequences, followed by motion tracking estimation using deterministic approaches. These included Lukas-Kanade Optical Flow and Simple Blob Detection, by OpenCV. The results showed that the grayscale image processing along with the Lukas-Kanade algorithm for motion tracking can produce better tracking abilities, although further exploration to improve the accuracy is still required.

Enhanced Object Detection in Autonomous Vehicles through LiDAR—Camera Sensor Fusion

To realize accurate environment perception, which is the technological key to enabling autonomous vehicles to interact with their external environments, it is primarily necessary to solve the issues of object detection and tracking in the vehicle-movement process. Multi-sensor fusion has become an essential process in efforts to overcome the shortcomings of individual sensor types and improve the efficiency and reliability of autonomous vehicles. This paper puts forward moving object detection and tracking methods based on LiDAR—camera fusion. Operating based on the calibration of the camera and LiDAR technology, this paper uses YOLO and PointPillars network models to perform object detection based on image and point cloud data. Then, a target box intersection-over-union (IoU) matching strategy, based on center-point distance probability and the improved Dempster–Shafer (D–S) theory, is used to perform class confidence fusion to obtain the final fusion detection result. In the process of moving object tracking, the DeepSORT algorithm is improved to address the issue of identity switching resulting from dynamic objects re-emerging after occlusion. An unscented Kalman filter is utilized to accurately predict the motion state of nonlinear objects, and object motion information is added to the IoU matching module to improve the matching accuracy in the data association process. Through self-collected data verification, the performances of fusion detection and tracking are judged to be significantly better than those of a single sensor. The evaluation indexes of the improved DeepSORT algorithm are 66% for MOTA and 79% for MOTP, which are, respectively, 10% and 5% higher than those of the original DeepSORT algorithm. The improved DeepSORT algorithm effectively solves the problem of tracking instability caused by the occlusion of moving objects.

A novel deep learning method for motion assessment in upper limb rehabilitation grasping test

Purpose The accuracy and reliability of upper limb motion assessment have received great attention in the field of rehabilitation. Grasping test is widely carried out for motion assessment, which requires patients to grasp objects and move them to target place. The traditional assessments test the upper limb motion ability by therapists, which mainly relies on experience and lacks quantitative indicators. This paper aims to propose a deep learning method based on the vision system of our upper limb rehabilitation robot to recognize the motion trajectory of rehabilitation target objects automatically and quantitatively assess the upper limb motion in the grasping test. Design/methodology/approach To begin with, an SRF network is designed to recognize rehabilitation target objects grasped in assessment tests. Moreover, the upper limb motion trajectory is calculated through the motion of objects’ central positions. After that, a GAE network is designed to analyze the motion trajectory which reflects the motion of upper limb. Finally, based on the upper limb rehabilitation exoskeleton platform, the upper limb motion assessment tests are carried out to show the accuracy of both object recognition of SRF network and motion assessment of GAE network. The results including object recognition, trajectory calculation and deviation assessment are given with details. Findings The performance of the proposed networks is validated by experiments that are developed on the upper limb rehabilitation robot. It is implemented by recognizing rehabilitation target objects, calculating the motion trajectory and grading the upper limb motion performance. It illustrates that the networks, including both object recognition and trajectory evaluation, can grade the upper limb motion functionn accurately, where the accuracy is above 95.0% in different grasping tests. Originality/value A novel assessment method of upper limb motion is proposed and verified. According to the experimental results, the accuracy can be remarkably enhanced, and the stability of the results can be improved, which provide more quantitative indicators for further application of upper limb motion assessment.

CTNeRF: Cross-time Transformer for dynamic neural radiance field from monocular video

The goal of our work is to generate high-quality novel views from monocular videos of complex and dynamic scenes. Prior methods, such as DynamicNeRF, have shown impressive performance by leveraging time-varying dynamic radiation fields. However, these methods have limitations when it comes to accurately modeling the motion of complex objects, which can lead to inaccurate and blurry renderings of details. To address this limitation, we propose a novel approach that builds upon a recent generalization NeRF, which aggregates nearby views onto new viewpoints. However, such methods are typically only effective for static scenes. To overcome this challenge, we introduce a module that operates in both the time and frequency domains to aggregate the features of object motion. This allows us to learn the relationship between frames and generate higher-quality images. Our experiments demonstrate significant improvements over state-of-the-art methods on dynamic scene datasets. Specifically, our approach outperforms existing methods in terms of both the accuracy and visual quality of the synthesized views. Our code is available on https://github.com/xingy038/CTNeRF.

DEVELOPMENT OF THE FACTUAL, CONCEPTUAL AND PROSEDURAL QUESTIONS ON THE MOVING MATERIALS OF CLASS VIII SMP

The students' knowledge of Physics Science material, especially in the subject of the motion of objects, is still below the expected KKM standard. Motion material is the basis or prerequisite for understanding further science material related to object motion material in class VIII to class IX. So far, there has never been an assessment related to the abilities of students' knowledge dimensions. This research focuses on assessing faktual, konseptual and prosedural knowledge according to the IC issued by the Ministry of Education and Culture. This type of research is development research using the ADDIE (Analyze, Design, Development, Implementation, Evaluation) development model. The purpose of this research is to describe the process of developing factual, conceptual and prosedural knowledge questions and to find out the dimensions of students' knowledge on the material motion of objects. The data obtained in this research is in the form of qualitative and quantitative data. Empirical validity analysis of 14 multiple-choice items and essays yielded the characteristics of the questions consisting of validity of 13 (92.8%) valid items, reliability of items with Cronbach's alpha 0.630 on 8 multiple-choice items and Cronbach's alpha 0.638 on 5 essay items greater than the r table value of 0.6, the level of difficulty is 64,3% items with moderate criteria, the discriminating power is 28,57% very good, 28,57% good questions, and 28,57% sufficient.

Pengolahan Citra Berbasis Video Proccesing dengan Metode Frame Difference untuk Deteksi Gerak

This study discusses the detection of motion of objects in video by utilizing the Frame difference method which aims to process video so as to produce Frames on moving objects. The use of mobile cameras produces video data that is used as test data, the test data is processed with the Frame difference method so as to produce a number of Frames on moving objects in order to detect moving objects in the video because the function of this method is a form of video background reduction that is simplified by a number of pixels in the video. This method process is based on the difference between two consecutive frames in the video aimed at finding differences that occur during the detection process. When processed for detection, the absolute value in the pixel is greater than the predetermined threshold value, it will be considered as a moving object, so that the detection results from the motion detection process will form a box object on the moving object. In this study, the test data used used 20 video data samples with descriptions, 10 test data with bright quality (daytime) and 10 unlit test data (night) with the aim of being able to see how much the level of performance accuracy of the Frame difference method. The test results obtained 16 out of 20 test data that were successfully detected correctly (True Positive), there were 2 test data that resulted in a False Positive error, and 2 test data that resulted in a False Negative error. This shows that the Frame difference method can provide a fairly high level of accuracy in detecting moving objects in the video. The percentage level of accuracy with confussion matrix testing has a precission value of 88%, recal 88% and an accuracy value of 80%.

A dynamic scale-mixture model of motion in natural scenes.

Some of the most important tasks of visual and motor systems involve estimating the motion of objects and tracking them over time. Such systems evolved to meet the behavioral needs of the organism in its natural environment, and may therefore be adapted to the statistics of motion it is likely to encounter. By tracking the movement of individual points in movies of natural scenes, we begin to identify common properties of natural motion across scenes. As expected, objects in natural scenes move in a persistent fashion, with velocity correlations lasting hundreds of milliseconds. More subtly, but crucially, we find that the observed velocity distributions are heavy-tailed and can be modeled as a Gaussian scale-mixture. Extending this model to the time domain leads to a dynamic scale-mixture model, consisting of a Gaussian process multiplied by a positive scalar quantity with its own independent dynamics. Dynamic scaling of velocity arises naturally as a consequence of changes in object distance from the observer, and may approximate the effects of changes in other parameters governing the motion in a given scene. This modeling and estimation framework has implications for the neurobiology of sensory and motor systems, which need to cope with these fluctuations in scale in order to represent motion efficiently and drive fast and accurate tracking behavior.

Kalman Filter for a Particular Class of Dynamic Object Images

We discuss the problem of estimating the state of a dynamic object by using observed images generated by an optical system. The work aims to implement a novel approach that would ensure improved accuracy of dynamic object tracking using a sequence of images. We utilize a vector model that describes the object image as a limited number of vertexes (reference points). Upon imaging, the object of interest is assumed to be retained at the center of each frame, so that the motion parameters can be considered as projections onto the axes of a coordinate system matched with the camera's optical axis. The novelty of the approach is that the observed parameters (the distance along the optical axis and angular attitude) of the object are calculated using the coordinates of specified points in the object images. For estimating the object condition, a Kalman-Bucy filter is constructed on the assumption that the dynamic object motion is described by a set of equations for the translational motion of the center of mass along the optical axis and variations in the angular attitude relative to the image plane. The efficiency of the proposed method is illustrated by an example of estimating the object's angular attitude.

The Development of Hologram Media on the Motion of Object in Physics to Improve Students Learning Outcomes in Junior High School

The quality of science learning in Indonesia is still low based on PISA results. One way to improve the quality of learning is to develop innovative learning media. This study aims to produce a supporting book and hologram animated videos as innovative media for learning science on the motion of objects in physics. Products are developed with the ADDIE model and then tested for their effectiveness. According to the results of the expert validation test for hologram animated video and a supporting book, each product obtains very feasible criteria. Furthermore, the results of user responses test for hologram animated videos, a supporting book, and their ability as innovative media according to users also get very feasible criteria. Then the result of the effectiveness test, the average value for the pre-test was 52.83 and the post-test was 82.17. These values are normally distributed by the Shapiro-Wilk test and had a significant difference by the paired sample t-test. In addition, the N-Gain test, a percentage of 62.10% is obtained with quite effective criteria. Thus, it can be concluded that our hologram media on the motion of objects in physics is very feasible to use and quite effective in improving student learning outcomes.

Research on the hydraulic parameters and leak rate of natural gas flow in natural gas pipelines based on the methods of characteristic line and successive approximation

Pressure and flow rate are the most important hydraulic parameters in natural gas pipeline flow, and leak rate is the most crucial parameter after a leak accident occurs. The study of these parameters is vital to the safey of natural gas pipelines and risk assessment after accidents. In this research, based on the conservation laws universally applicable to the motion of objects, we establish the fundamental control equation group for natural gas flow. Then, for both normal and leak conditions of pipelines, we use the characteristic line method to derive the corresponding difference equations for the fundamental control equations, thereby providing calculation methods for pressure and flow rate. Finally, we investigate the calculation method of natural gas pipeline leak rate under different leak aperture sizes, and validate the accuracy of this method through simulation examples.

High-quality direct ghost imaging of random dynamic targets based on convolutional neural network

We propose a novel high-quality direct random dynamic target ghost imaging method, based on Convolutional Neural Networks (CNNs). This method requires input of one-dimensional bucket detector values collected during the motion of the target object, enabling direct reconstruction of the target image. The results demonstrate the method’s capability to achieve high-quality reconstructed images for moving targets across various sampling rates. Furthermore, the method effectively addresses the challenge of image quality degradation caused by target motion. This contributes to the potential practical application value of ghost imaging in fields such as remote sensing and radar.

The complete exterior spacetime of spherical Brans-Dicke stars

We derive the complete expression for the Brans Class I exterior spacetime explicitly in terms of the energy and pressures profiles of a stationary spherisymmetric gravity source. This novel and generic expression is achieved in a parsimonious manner, requiring only a subset of the Brans-Dicke field equation and the scalar equation. For distant orbiting test particles, this expression promptly provides a simple, closed and exact formula of the γ Eddington parameter, which reads γexact=ω+1+(ω+2)Θω+2+(ω+1)Θ, where Θ is the ratio of the star's “total pressure” integral over its energy integral. This non-perturbative result reproduces the usual Post-Newtonian ω+1ω+2 expression in the case of a “Newtonian star”, in which the pressure is negligible with respect to the energy density. Furthermore, it converges to the General Relativity value (γGR=1) as the star's equation of state approaches that of ultra-relativistic matter (in which case Θ approaches 1), a behavior consistent with broader studies on scalar-tensor gravity. Our derivation underscores the essence of these results involving (1) the key relevant portion of the Brans-Dicke field equations, (2) the uniqueness of the Brans Class I vacuum solution for the non-phantom action, viz. ω>−3/2, and (3) the involvement of only two free parameters in this solution, hence requiring two quantities (energy and pressure integrals) of the mass source to fully characterize the solution. From a practical standpoint, it elucidates how a given stellar interior structure model determines the star's exterior gravitational field and impacts the motions of light objects (such as planets and accretion disks) orbiting it.

Ellipsoidal estimation of motion parameters of a non-cooperative space vehicle from visual information

The use of near-Earth space is currently complicated by presence of space debris objects in Earth’s orbit, which include spent stages of launch vehicles, inoperative satellites, and other large and small artificial objects. One approach to solving the problem of space debris involves docking and capturing a non-cooperative space object or spacecraft (target) with a so-called service spacecraft (chaser) for further actions to repair, refuel or change its orbit. Rendezvous and docking are complicated by the rotation of uncontrolled space objects caused by various factors. To perform this task, it is necessary to know the parameters of the orbital, rotational and relative motion of the target. The parameters of the orbital motion of such objects are usually known quite accurately from measurements from the Earth. This paper examines the case of a tumbling non-cooperative target located in an elliptical orbit. It is assumed that the target relative position and orientation are measured by the computer vision system (CVS) of the chaser. In this case, the position and orientation of the graphical reference frame (GRF) associated with the known 3-D graphical model of the target are determined relative to the reference frame associated with the chaser. The specific type of CVS is not considered. It is assumed that the chaser can carry out some maneuvers near the target and all parameters of the chaser angular motion are known. Thus, the attitude of the GRF relative to inertial reference frame (IRF) can be determined. The measured parameters are not enough to ensure safe rendezvous and docking with the target. To complete this task, it is necessary to determine all kinematic and dynamic parameters of the relative motion between the spacecraft. The rest of the required parameters are estimated. Orientation and rotation parameterization is done using quaternions. The angular motion equation of the target is considered in the GRF. This makes the angular velocity estimation faster and the inertia tensor estimation more stable. Stochastic characteristics of measurement errors are considered to be unknown and are not used. The only information about the errors is the bounds of their values. To determine the relative motion parameters, we use a new dynamic set-membership filter with ellipsoidal estimates. The filter can be successfully implemented on low-power on-board processors. The properties of the proposed algorithm are demonstrated using numerical simulation. The results obtained are expected to be used in the development of a navigation system for the rendezvous and docking, developed by a group of Ukrainian space industry enterprises under the leadership of the LLC “Kurs-orbital” (https://kursorbital.com/).

Spatial-temporal graph Transformer for object tracking against noise spoofing interference

Achieving accurate object tracking against noise spoofing interference caused by similar backgrounds, similar objects, occlusion, or illumination variations is a challenge, especially since Siamese trackers largely ignore the spatial-temporal and part-level information of object structure. To handle this problem, we present a novel object tracking network with spatial-temporal awareness, which learns the correlation of local feature changes and extracts global spatial-temporal fusion features of the object to prevent noise spoofing interference. Specifically, our tracker integrates an object graph reconstruction representation module and a spatial-temporal graph Transformer module. The graph reconstruction representation module models the object structure and search region as part-to-part graph node correspondences, propagating object information to achieve part-level feature aggregation in the search region. The proposed spatial-temporal graph Transformer module fuses the temporal and spatial features of the object to enhance the contextual features and correlation between temporal and spatial feature variations, recognizing the noise spoofing interference to enhance the tracking performance. Meanwhile, our tracker learns object motion information to improve object state awareness, enhancing tracking accuracy under similar objects interference. Extensive experiments on six public datasets validate that our tracker outperforms related state-of-the-art trackers and achieves accurate tracking results against noise spoofing interference.

Self-powered noncontact triboelectric nanogenerators with microstructured square-loop surface and dielectric electron-blocking layer for far-distance motion perception and trajectory tracking

Continuous monitoring of indoor human activities is crucial for the healthcare of elder people or patients. The self-powered noncontact triboelectric nanogenerator (TENG) sensor has been recently developed as a promising noncontact sensing tool with many advantages, but its capability of detecting object motion in a far distance still remains inferior. Herein, by implementing multiple strategies of micro-structuring triboelectric surface, introducing dielectric layer, and using functional nano fillers, we develop a single-electrode-mode noncontact TENG sensor by constructing a triboelectric cobalt nanoporous carbon/Ecoflex nanocomposite layer with microstructured square-loop surface on top of a dielectric copper calcium titanate/Ecoflex nanocomposite layer with colossal permittivity. After in-depth optimization of the material selection, component composition and structure design, the developed sensor with effective electrode area of 6.25 cm2 can deliver a high output voltage of 29 V at 1 cm, and is capable of detecting motions at an extremely far distance of 6 m. Towards practical applications, the TENG unit for single-motion detection of mouth breathing, human walking, and basketball bouncing are demonstrated. Furthermore, multiple TENG units can be assembled into array to realize the complex trajectory tracking for the use of proximity touch pad, indoor human activity monitoring, and garage car parking. This work for the first time extends the TENG noncontact sensing capability to several meters, which is potential in the proximity motion detection and indoor activity monitoring.

Self-supervised rigid object 3-D motion estimation from monocular video

Independent object 3D motion estimation is a fundamental problem in 3D computer vision. Directly segmenting and estimating rigid object 3D motion from a consistent video frame is an ill-posed problem. We present a self-supervised framework for segmenting moving independent rigid objects and estimating their motion information including location, driving direction, and speed from a monocular video. Specifically, we first estimate depth, optical flow, and camera pose between a pair of video frames, and then synthesize a new 3D viewpoint from this pair. Subsequently, the Motion Recurrent All-Pairs Field Transforms (MRAFT) module is introduced to extract 3D scene flow and a motion area binary mask from a pair of images and depth. After that, a Rigid Object Motion Estimation Module (ROMEM) with a slot attention mechanism is proposed to extract rigid object motion masks from a multi-layer motion field, including optical flow, depth changes, refined scene flow, and motion masks. Finally, 2D images and 3D scene reconstruction errors are used to facilitate self-supervised training for rigid object motion. Experiments on the FlyingThings3D and KITTI datasets demonstrate that our method outperforms other advanced algorithms in estimating depth, optical flow, scene flow, and rigid moving object masks, demonstrating the benefits of our approach.

Research on Object Motion, Fire and Gas Detection Using ESP32

Abstract: Working parents frequently worry about leaving their young children unattended at home, but thanks to technological advancements, it is now simpler to keep an eye on things and send information quickly. With the aid of the Internet of Things, parents may simply monitor their children's behavior within the home or keep an eye on guests at the entrance. By sensing motion and sending photographs on Telegram, this hardware-based application seeks to improve home security and kid safety by assisting in identifying any anomalous activity. This hardware-based architecture saves time and aids in monitoring in lieu of the owner. The fast expansion of industrial facilities and infrastructure is causing pollution, climate change, and other environmental problems. The suggested device uses humidity sensors, temperature detectors (DHT 11), gas detectors (MQ2, MQ7, and MQ135),and other sensors to identify the unfavorable gases present in the production process.

Avian eye-inspired perovskite artificial vision system for foveated and multispectral imaging.

Avian eyes have deep central foveae as a result of extensive evolution. Deep foveae efficiently refract incident light, creating a magnified image of the target object and making it easier to track object motion. These features are essential for detecting and tracking remote objects in dynamic environments. Furthermore, avian eyes respond to a wide spectrum of light, including visible and ultraviolet light, allowing them to efficiently distinguish the target object from complex backgrounds. Despite notable advances in artificial vision systems that mimic animal vision, the exceptional object detection and targeting capabilities of avian eyes via foveated and multispectral imaging remain underexplored. Here, we present an artificial vision system that capitalizes on these aspects of avian vision. We introduce an artificial fovea and vertically stacked perovskite photodetector arrays whose designs were optimized by theoretical simulations for the demonstration of foveated and multispectral imaging. The artificial vision system successfully identifies colored and mixed-color objects and detects remote objects through foveated imaging. The potential for use in uncrewed aerial vehicles that need to detect, track, and recognize distant targets in dynamic environments is also discussed. Our avian eye-inspired perovskite artificial vision system marks a notable advance in bioinspired artificial visions.

Secular structure of 1:2 and 1:3 mean motion resonances with Neptune

The 1:N mean motion resonances (MMRs) with Neptune are of particular interest in astrophysical research because they have two asymmetric resonance islands, where the distribution of trapped objects may bear important clues to resolving the history of the Solar System. To explore the dynamics of these resonances and to investigate whether the imprints left by the early stage evolution can be preserved in the resonances, we conducted a comprehensive analysis of the 1:2 and 1:3 resonances. By mainly adopting the frequency analysis method, we calculated the proper frequencies of the motion of objects in the resonances and determined the secular mechanisms that influence the dynamics. Using the spectral number (SN) as an indicator of orbital regularity, we constructed dynamical maps on representative planes. After comparing the structures in the maps with the locations of the secular mechanisms, we find that the von-Zeipel-Lidov-Kozai mechanism and the $g=2s$ mechanism destabilize the influenced orbits and thus sculpt the overall structure of the 1:2 and 1:3 resonances. The secular resonance of $2g-s=s_8$ opens a channel for objects to switch between the leading and trailing resonance islands, which can alter the population ratio between these two islands. The secondary resonances associated with the quasi 2:1 resonance between Uranus and Neptune were also detected, likely introducing more chaos to the motion. The fine dynamical structures of the 1:2 and 1:3 resonances revealed in this paper, combined with knowledge of resonant capture, provide a compelling explanation for the eccentricity distribution of observed Twotinos. Furthermore, we anticipate a more complete understanding of the history of planetary migration in the Solar System can be achieved by comparing the results in this paper with the populations in the 1:N resonances, with forthcoming observations offering more objects for study in the future.