Multiple Cameras Research Articles

Laboratory Spectral Characterization of Ribbeck Aubrite: Meteorite Sample of Earth-impacting Near-Earth Asteroid 2024 BX1

Characterization of near-Earth objects (NEOs) is critical for Earth-impact hazard assessment. Particularly crucial to our physical understanding of NEOs are laboratory spectral measurements of meteorites as they are the best and most widely available analog materials, barring sample return missions. However, most meteorites do not have direct orbital links to specific asteroids, making it challenging to identify their source body in the NEO or main-belt asteroid populations. Near-Earth asteroid (NEA) 2024 BX1 was discovered on 2024 January 20 at 21:48 UTC from MPC code K88, impacting the Earth (west of Berlin, Germany) 165 minutes later. The incoming bolide was observed by multiple meteor cameras, which enabled successful reconstruction of its exo-atmospheric orbit and quick recovery. We present results from laboratory spectral characterization of the Ribbeck meteorite in the UV–mid-infrared wavelengths (0.2–14.2 μm) over seven grain size bins (<45 μm–slab). Our results suggest that Ribbeck has spectral properties consistent with enstatite achondrite (aubrite) meteorites. Our grain-size spectral analysis shows that albedo and spectral slope decrease as grain size increases. In addition, increasing grain size also shifts the taxonomic type in the Bus–DeMeo system from Xn to B types, suggesting the limitations of taxonomy in classifying small, regolith-free NEAs. We also present results of our comparison between Ribbeck data and spectra of E types in the main-belt and NEA populations. Principal component analysis of our Ribbeck samples shows variations parallel to the α line, which can be confused with space weathering in PC space.

Research and Application on High Precision Measurement Method of Large Mesh Antenna Surface

Abstract The development process of large mesh antennas requires a significant investment of time and resources for iterative measurements and adjustments to achieve high precision in the large mesh reflector. In this paper, a novel measurement method for the precision of large mesh antenna surfaces is proposed to reduce the development process. Multiple measurement cameras have been utilized to set up a network system, enabling data integration for comprehensive analysis and computational calculations. In this approach, the implementation of a one-click and the rapid measurement function represent substantial advancements. The single measurement time has been successfully reduced from 40 minutes to 2 minutes, which addresses the challenge of high efficiency and precision measurement application on the complex mesh antenna surface. Then the optimization of calibration is accomplished through the meticulous measurement of network patterns, marker points, and reference bars to improve the precision accuracy on the foundation of rapid measurements, combined with the simulation analysis. The final results have been practically verified and successfully applied to model development, which provides a crucial reference for subsequent endeavors involving the rapid and automated measurement of similar antennas.

Deep learning based lily bulb classification from multiple camera views

Deep learning based lily bulb classification from multiple camera views

3-Dimensional Biomechanics of Noncontact Anterior Cruciate Ligament Injuries in Male Professional Soccer Players

Background: The understanding of noncontact anterior cruciate ligament (ACL) injury causation in soccer has improved over the past decades. Bidimensional video analyses have significantly augmented our awareness, representing to date the only practical method to describe injury biomechanics. However, the extent of the problem continues to raise serious concerns. Purpose: To advance our understanding of the causal pathways leading to ACL injury with a large-scale reconstruction of 3-dimensional (3D) whole-body joint kinematics of injuries that occurred to male elite soccer players, as well as to compare the joint angle time course among situational patterns. Study Design: Descriptive laboratory study. Methods: A total of 33 consecutive noncontact and indirect contact ACL injuries that occurred in 6 national and 2 international professional leagues (seasons 2020-2021 to 2022-2023 until December 2022) were analyzed: (1) multiview noncoaxial television images were inspected; (2) multiple camera views were taken from 400 ms before the initial ground contact to 200 ms after the injury frame; (3) a size-matched pitch was modeled and used to calibrate cameras; (4) a 3D skeletal model was adjusted to fit the player's pose in each frame/view; and (5) poses were interpolated, and Euler joint angles were extracted. Results: The authors reconstructed the 3D lower limb joint kinematic curves preceding and during ACL injuries in 33 cases; notably, a sudden external (up to 5°) and then internal knee rotation was observed after the initial contact and before the injury frame. The overall kinematics at injury were knee moderately flexed (45.9°± 21.7°), abducted (4.3°± 5.1°), and externally rotated (3.0°± 6.4°); trunk shallowly flexed (17.4°± 12.5°) and rotated and tilted toward the injured side; and hip flexed (32.0°± 18.7°), abducted (31.1°± 12.0°), and slightly internally rotated (6.6°± 12.2°). Variable behaviors were observed at the ankle level. Conclusion: Via reconstruction of the sequence of whole-body joint motion leading to injury, we confirmed the accepted gross biomechanics (dynamic valgus trend). This study significantly enriches the current knowledge on multiplanar kinematic features (transverse and coronal plane rotations). Furthermore, it was shown that ACL injuries in male professional soccer players manifest through distinct biomechanical footprints related to the concurrent game situation. Clinical Relevance: Interventions aimed at reducing ACL injuries in soccer should consider that environmental features (ie, situational patterns) affect injury mechanics.

Track initialization and re-identification for 3D multi-view multi-object tracking

We propose a 3D multi-object tracking (MOT) solution using only 2D detections from monocular cameras, which automatically initiates/terminates tracks as well as resolves track appearance–reappearance and occlusions. Moreover, this approach does not require detector retraining when cameras are reconfigured but only the camera matrices of reconfigured cameras need to be updated. Our approach is based on a Bayesian multi-object formulation that integrates track initiation/termination, re-identification, occlusion handling, and data association into a single Bayes filtering recursion. However, the exact filter that utilizes all these functionalities is numerically intractable due to the exponentially growing number of terms in the (multi-object) filtering density, while existing approximations trade-off some of these functionalities for speed. To this end, we develop a more efficient approximation suitable for online MOT by incorporating object features and kinematics into the measurement model, which improves data association and subsequently reduces the number of terms. Specifically, we exploit the 2D detections and extracted features from multiple cameras to provide a better approximation of the multi-object filtering density to realize the track initiation/termination and re-identification functionalities. Further, incorporating a tractable geometric occlusion model based on 2D projections of 3D objects on the camera planes realizes the occlusion handling functionality of the filter. Evaluation of the proposed solution on challenging datasets demonstrates significant improvements and robustness when camera configurations change on-the-fly, compared to existing multi-view MOT solutions.

How the Science Behind Video Games Helps Kids Walk

Did you know that some of the same technology used to make video games and movie special effects can help scientists better understand human movement? During motion capture, also called mocap, small reflective balls, or markers as they are called, are placed on specific points of a person’s body. Mocap systems work using multiple cameras around the room that track the motion of the markers as a person walks, runs, jumps, or plays. The information gathered during mocap can be used to help doctors decide the best possible care to improve a child’s walking. It can also be used to determine when someone is ready to go back to sports after an injury, to help prevent people from getting injuries in the first place, or to improve their sports performance.

Underwater Camera Calibration Based on Double Refraction

Underwater camera calibration plays a pivotal role in underwater positioning and underwater mapping reconstruction, making it crucial for achieving precise spatial measurements in underwater environments. To elevate measurement accuracy, we have refined the calibration methodology for underwater cameras. Firstly, we conducted an in-depth investigation into the intricate challenges posed by double refraction errors arising from light passing through water–glass–air interfaces. To address this issue, we established a double refraction model based on the actual underwater light paths, laying the groundwork for our calibration efforts. Furthermore, to tackle the problem of determining multiple camera parameters, we proposed a parameter optimization method based on genetic algorithms, capable of navigating the complex parameter space. Finally, compared with other algorithms, our method enables more precise determination of underwater camera parameters.

Optimizing Camera Motion with MCTS and Target Motion Modeling in Multi-Target Active Object Tracking

In this work, we are dedicated to multi-target active object tracking (AOT), where the goal is to achieve continuous tracking of targets through real-time control of camera. This form of active camera control can be applied to unmanned aerial vehicles (UAV), intelligent robots, and sports events. Our work is conducted in an environment featuring multiple cameras and targets, where our goal is to maximize target coverage. Contrasting with previous research, our work introduces additional degrees of freedom for the cameras, allowing them not only to rotate but also to move along boundary lines. In addition, we model the motion of target to predict the future position of the target in environment. With target’s future position, we use Monte Carlo Tree Search (MCTS) method to find the optimal action of camera. Since the action space is large, we propose to leverage the action selection from multi-agent reinforcement learning (MARL) network to prune the search tree of Monte Carlo Tree Search method, so as to find the optimal action more efficiently. We establish a multi-target 2D environment to simulate several sports games, and experimental results demonstrate that our method can effectively improve the target coverage. The code is available at: http://github.com/HopeChanger/ActiveObjectTracking .

Multiscale 3D displacement field measurement using stereo digital image correlation on a fractal speckle pattern

AbstractEven though the simulations used to predict failure are becoming increasingly predictive, complex multiaxial loading tests are still required to validate the design of structural components in a wide range of industries. Large specimen testing often requires two different scales. A global Far Field to obtain boundary conditions and a local Near Field to evaluate strain gradients around discontinuities such as bolts, notches… The main goal of this study is to provide a continuous displacement over the whole specimen surface integrating data from multiple cameras. In this paper, we propose a new methodology that generates 3D displacements determined by finite‐element stereo digital image correlation in the Near Field and in the Far Field using a unique fractal speckle pattern and an off‐line determined texture. The displacements are obtained in the same coordinate system and on the same mesh. Satisfactory data fusion from both Near Field and Far Field images of a biaxial test on a notched laminate composite was obtained with a refined mesh at the notch tip. This methodology can be applied to any tests requiring multiple camera systems and will support the use of the finite‐element digital image correlation framework as an experimental‐numerical efficient technique.

Deep video-based person re-identification (Deep Vid-ReID): comprehensive survey

Person re-identification (ReID) aims to find the person of interest across multiple non-overlapping cameras. It is considered an essential step for person tracking applications which is vital for surveillance. Person ReID could be investigated either using image-based or video-based. Video-based person ReID is considered more discriminating and realistic than image-based ReID due to the massive information extracted for each person. Different deep-learning techniques have been used for video-based ReID. In this survey, recently published articles are reviewed according to video-based ReID system pipeline: deep features learning, deep metric learning, and deep learning approaches. The deep feature learning approaches are categorized into spatial and temporal approaches, while deep metric learning is divided into metric and metric learning approaches. The deep learning approaches are differentiated into: supervised, unsupervised, weakly-supervised, and one-shot learning. A detailed analysis is held for the architectures of the state-of-the-art deep learning approaches. And their performance on four benchmark datasets is compared.

Realistic Texture Mapping of 3D Medical Models Using RGBD Camera for Mixed Reality Applications

Augmented and mixed reality in the medical field is becoming increasingly important. The creation and visualization of digital models similar to reality could be a great help to increase the user experience during augmented or mixed reality activities like surgical planning and educational, training and testing phases of medical students. This study introduces a technique for enhancing a 3D digital model reconstructed from cone-beam computed tomography images with its real coloured texture using an Intel D435 RGBD camera. This method is based on iteratively projecting the two models onto a 2D plane, identifying their contours and then minimizing the distance between them. Finally, the coloured digital models were displayed in mixed reality through a Microsoft HoloLens 2 and an application to interact with them using hand gestures was developed. The registration error between the two 3D models evaluated using 30,000 random points indicates values of: 1.1 ± 1.3 mm on the x-axis, 0.7 ± 0.8 mm on the y-axis, and 0.9 ± 1.2 mm on the z-axis. This result was achieved in three iterations, starting from an average registration error on the three axes of 1.4 mm to reach 0.9 mm. The heatmap created to visualize the spatial distribution of the error shows how it is uniformly distributed over the surface of the pointcloud obtained with the RGBD camera, except for some areas of the nose and ears where the registration error tends to increase. The obtained results indicate that the proposed methodology seems effective. In addition, since the used RGBD camera is inexpensive, future approaches based on the simultaneous use of multiple cameras could further improve the results. Finally, the augmented reality visualization of the obtained result is innovative and could provide support in all those cases where the visualization of three-dimensional medical models is necessary.

Assessing Spatiotemporal Behavior of Human Gait: A Comparative Study Between Low-Cost Smartphone-Based Mocap and OptiTrack Systems

AbstractThis study assessed the accuracy of a low-cost marker-based motion capture system with smartphone devices to estimate the spatiotemporal behavior of human gait in comparison with the performance of the commercial OptiTrack system. Initially, three test subjects were selected for the study, and after equipping them with passive retroreflective markers, they were recorded for gait velocities of 1.50, 1.90, and 2.30 $$m\bullet {s}^{-1}$$ m ∙ s - 1 while collecting kinematic data and videos. The results showed that the smartphone motion capture system exhibited significant spatiotemporal tracking and accuracy in the x-y trajectories and estimation of joint relative angles of the hip, knee, and ankle joints (θ1, θ2, and θ3, respectively) compared to the commercial OptiTrack system. In this comparison, an average goodness-of-FIT and normalized root mean square error of over 88.93% and 2.71% were obtained, respectively, for the joint relative angles of the hip and knee (θ1 and θ2) in all tests performed. However, the accuracy of the joint relative angle of the ankle (θ3, average FIT: 71.04% and nRMSE: 4.26%) was lower because of the low capture rate of the retroreflective markers in the smartphone system and the higher relative velocity in the lower extremities of the test subjects, which generated noise in the calculation of x-y trajectories. This decrease in accuracy has been reported in other studies. However, both motion capture systems experienced marker data loss at the hip, highlighting the need for improvement in the spatial distribution of the optical devices. The OptiTrack system demonstrated better optical redundancy but still required improvements. In contrast, the smartphone system, with its inherent limitations in terms of optical redundancy and spatial distribution, can be enhanced by incorporating multiple cameras for a three-dimensional view. Despite these limitations, the low-cost smartphone system showed optimal performance with minimal errors compared with the commercial system, making it a cost-effective option with potential for further development. The rapid advancement of smartphone technology and its accessibility make it an attractive choice for motion capture applications.

Real-Time Deformation Measurement of Long-Span Bridge using Multi-Inertial Camera System

Abstract. Monitoring the deformation of long-span bridges is essential for assessing structural health and safety. The existing single-camera deformation measurement method is unable to meet the high-precision measurement requirement for the deformation of large-span bridges. This paper proposes a real-time deformation measurement method for the long-span bridge using multi-inertial camera system. The method is based on visual measurement which the target deformation is measured by multiple cameras. Meanwhile, inertial measurement is used to estimate the relative pose of the camera to compensate for camera motion errors. It is applied to the health monitoring system of a long-span bridge. The experimental results show that the method proposed in this paper is highly consistent with the hydrostatic leveling measurement, with a RMSE of 4.83mm. The method can accurately and real-time measure the deformation of large-span bridges with a promising engineering value.

Simultaneous Calibration of Multiple Cameras and Generation of Omnidirectional Images

Abstract. Omnidirectional images are increasingly being used in various areas, such as urban mapping, virtual reality, agriculture, and robotics. These images can be generated by different acquisition systems, including multi-camera systems, which can acquire higher-resolution images. Stitching techniques are often used and can be suitable for non-metric applications, but rigorous photogrammetric processing is recommended when having more accurate requirements. The main challenges related to this kind of product are the system calibration and the generation of the final omnidirectional images. When using multi-camera systems, the displacement of the cameras' perspective centres can affect the generation of the omnidirectional images and the resulting accuracy. A common approach to minimising the resulting parallax error is to establish a value for the projection cylinder radius as close as possible to the object's depth. This work proposes a highly accurate simultaneous calibration technique for multiple camera systems using self-calibrating bundle adjustment with constraints of stability of the relative orientation parameters. These parameters are later used to generate a projecting cylindrical surface, maintaining the original camera perspective centres and relative orientation angles. The experiments show that using constraints improved both the calibration results and the final omnidirectional images. Residual mismatches between points in overlapping areas are subpixel.

3D-MuPPET: 3D Multi-Pigeon Pose Estimation and Tracking

Markerless methods for animal posture tracking have been rapidly developing recently, but frameworks and benchmarks for tracking large animal groups in 3D are still lacking. To overcome this gap in the literature, we present 3D-MuPPET, a framework to estimate and track 3D poses of up to 10 pigeons at interactive speed using multiple camera views. We train a pose estimator to infer 2D keypoints and bounding boxes of multiple pigeons, then triangulate the keypoints to 3D. For identity matching of individuals in all views, we first dynamically match 2D detections to global identities in the first frame, then use a 2D tracker to maintain IDs across views in subsequent frames. We achieve comparable accuracy to a state of the art 3D pose estimator in terms of median error and Percentage of Correct Keypoints. Additionally, we benchmark the inference speed of 3D-MuPPET, with up to 9.45 fps in 2D and 1.89 fps in 3D, and perform quantitative tracking evaluation, which yields encouraging results. Finally, we showcase two novel applications for 3D-MuPPET. First, we train a model with data of single pigeons and achieve comparable results in 2D and 3D posture estimation for up to 5 pigeons. Second, we show that 3D-MuPPET also works in outdoors without additional annotations from natural environments. Both use cases simplify the domain shift to new species and environments, largely reducing annotation effort needed for 3D posture tracking. To the best of our knowledge we are the first to present a framework for 2D/3D animal posture and trajectory tracking that works in both indoor and outdoor environments for up to 10 individuals. We hope that the framework can open up new opportunities in studying animal collective behaviour and encourages further developments in 3D multi-animal posture tracking.

Real-Time Seamless Multi-Projector Displays on Deformable Surfaces.

Prior works on multi-projector displays have focused primarily on static rigid objects, some focusing on dynamic rigid objects. However, works on projection based displays on deformable dynamic objects have focused only on small scale single projector displays. Tracking a deformable dynamic surface and updating projections precisely in real time on it is a significantly challenging task, even for a single projector system. In this paper, we present the first end-to-end solution for achieving a real-time, seamless display on deformable surfaces using mutliple unsychronized projectors without requiring any prior knowledge of the surface or device parameters. The system first accurately calibrates multiple RGB-D cameras and projectors using the deformable display surface itself, and then using those calibrated devices, tracks the continuous changes in the surface shape. Based on the deformation and projector calibration, the system warps and blends the image content in real-time to create a seamless display on a surface that continuously changes shape. Using multiple projectors and RGB-D cameras, we provide the much desired aspect of scale to the displays on deformable surfaces. Most prior dynamic multi-projector systems assume rigid objects and depend critically on the constancy of surface normals and non-existence of local shape deformations. These assumptions break in deformable surfaces making prior techniques inapplicable. Point-based correspondences become inadequate for calibration, exacerbated with no synchronization between the projectors. A few works address non-rigid objects with several restrictions like targeting semi-deformable surfaces (e.g. human face), or using single coaxial (optically aligned) projector-camera pairs, or temporally synchronized cameras. We break loose from such restrictions and handle multiple projector systems for dynamic deformable fabric-like objects using temporally unsynchronized devices. We devise novel methods using ray and plane-based constraints imposed by the pinhole camera model to address these issues and design new blending methods dependent on 3D distances suitable for deformable surfaces. Finally, unlike all prior work with rigid dynamic surfaces that use a single RGB-D camera, we devise a method that involve all RGB-D cameras for tracking since the surface is not seen completely by a single camera. These methods enable a seamless display at scale in the presence of continuous movements and deformations. This work has tremendous applications on mobile and expeditionary systems where environmentals (e.g. wind, vibrations, suction) cannot be avoided. One can create large displays on tent walls in remote, austere military or emergency operations in minutes to support large scale command and control, mission rehearsal or training operations. It can be used to create displays on mobile and inflatable objects for tradeshows/events and touring edutainment applications.

Indoor 3D Visible Light Positioning Based on Multiple Cameras: Algorithm Design and Error Analysis

Indoor 3D Visible Light Positioning Based on Multiple Cameras: Algorithm Design and Error Analysis

Large-scale flow field super-resolution via local-global fusion convolutional neural networks

Particle image velocimetry (PIV) techniques have a limited field of view of the flow field and can only capture high-resolution flow fields in localized areas. To obtain a larger measurement range, multiple cameras must be used to capture the flow field simultaneously and then stitch the parts together. However, this method can be expensive. We propose the local-global fusion convolutional neural network (LGF-CNN) for reconstructing large-field flow fields with high spatial resolution based only on two flow data types: local small-field high spatial resolution wake velocity fields and global large-field low spatial resolution velocity fields. The core of the model consists of convolutional neural network (CNN) architecture to learn the mapping relationship between the small field of view with high spatial resolution and the large field of view with low spatial resolution. Using the effectively trained LGF-CNN model, we demonstrate its ability to reconstruct high-resolution velocity fields around the circular cylinder. The LGF-CNN is rigorously validated on a number of representative datasets, including simulated data for Reynolds numbers of 200 and 500, as well as experimental data for a Reynolds number of 3.3 × 104 with the steady jet at the rear stagnation point of the cylinder. The results demonstrate the ability of LGF-CNN to generate accurate velocity fields with high spatial resolution, including reliable detection of high-frequency components. The proposed method could reduce the number of cameras required for large-field, high spatial resolution PIV measurements, thereby reducing experimental costs.

MobileNetV2-D and multiple cameras for swiftlet nest classification based on feather intensity

MobileNetV2-D is a modified version of MobileNetV2, which is the novelty of this article. The algorithm is used to classify swiftlet nests into seven classes. In 2023, PT Waleta Asia Jaya is required to achieve a 7-fold increase in the export quota of swiftlet nests. To meet the quota, the company made a machine that can recognize swiftlet nest objects, which are classified into seven classes based on feather intensity, namely BRS, BR, BST, BS, BBT, BB, and BB2 for the light feathers to the heavy feathers, respectively. The input image is a combination of four images from four cameras with different positions, which adds to the novelty of MobileNetV2-D for the particular problem here. From the evaluation that has been carried out, the accuracy value of the MobileNetV2-D model was better than the MobileNetV2 model, i.e., the accuracy value of the MobileNetV2-D model was 99.9928% for the training dataset and 94.0723% for the testing dataset. Moreover, the speed of MobileNetV2-D is better than MobileNetV2- architecture.

Investigation of the three-dimensional structure of low-swirl methane-hydrogen flames with multiple UV cameras-based OH* chemiluminescence tomography

Blending hydrogen with natural gas is useful for reducing CO2 emissions, but this alters the flame structure, creating challenges for flame stability. Numerous researches on methane-hydrogen flame structure have been carried out based on two-dimensional combustion diagnostic techniques such as flame chemiluminescence and planar laser-induced fluorescence. In this paper, a low-cost OH* chemiluminescence imaging system consisting of 8 ultraviolet (UV) cameras is proposed to reconstruct the three-dimensional (3D) structure of methane-hydrogen flames, and validated through use of the correlation coefficient. Further experimental investigations of the low-swirl methane-hydrogen flames are performed by this computed tomography of chemiluminescence (CTC) system. The flame properties are determined to evaluate the effects of various fractions of blended hydrogen on the flame structure, including the flame center of mass, flame surface density and fractal dimension. Results showed that the flame initially has bowl-shape structure. The flame height gradually decreases and the flame cross-sectional area increases as the blended hydrogen fraction increases. Meanwhile, the bottom of the flame moves toward the nozzle outlet. Nevertheless, when the blended hydrogen fraction reaches 50%, the flame transforms into a cup-shaped flame and shows a double-layer flame structure. When the flame is bowl-shaped, the flame center of mass moves away from the nozzle center axis with increasing blended hydrogen fraction, but the fractal dimension and flame surface density are not sensitive to the blended hydrogen fraction. In contrast, when the flame is cup-shaped, the flame center of mass moves towards the nozzle center axis, and the flame surface density and fractal dimension increase significantly, indicating the increase of flame wrinkling. These results reveal that blending hydrogen leads to two distinct flame structures and characteristics in low-swirl flame, also demonstrate that the proposed 3D low-cost CTC system based on multiple UV cameras is well-suited for combustion diagnostics.