Orientation Estimation Research Articles

Best practices towards the digitization of 3D traces from virtual outcrop models

With the expanding application of virtual outcrop models (VOMs) within the geosciences, there is a growing need to establish standards for the digital measurement of geological discontinuities exposed therein. Such standards should be tailored towards the complexities of natural outcrops, where geological discontinuities often intersect the outcrop topography and are expressed as 3D traces. Digitizing geological discontinuities expressed as trace data is typically conducted manually via polyline interpolation along the exposed trace, with discontinuity orientation estimated through planar model fitting through the polyline's component nodes. Presently, establishing quality control for such measurements lacks standardization due to the absence of robust benchmarks, with the validity of the resultant orientation data heavily reliant on the experience of the interpreter.With the aim of bridging this gap, we present the results of the manual digitization and orientation estimation of bedding planes expressed as traces across seven natural outcrops. We use two digitization strategies: one employing a previewed best-fit plane during digitization and another without. The first digitization method is carried out by an expert user who visually filters data according to visual alignment with the intended bedding prior to best fit plane estimation. In contrast, the non-visually aided method mimics acquisition by a novice user, with no a priori data filtering based upon trace geometry with respect to the outcrop. Comparison of the results obtained by these ‘expert’ and ‘novice’ acquisition modes is aimed at building benchmarks and best practices. Specifically, we analyze parameters derived from the digitized traces and their corresponding best-fit planes. We compare these parameters with the deviation of the best-fit plane from the mean orientation of the bedding surface as measured using visually-aided acquisition. Comparing these datasets reveals that visually-aided digitization yields more precise and accurate bedding measurements, characterized by traces with lower vertex collinearity. Notably, comparable results can be achieved in the non-visually assisted dataset by excluding traces with high node collinearity. Consequently, we provide robust benchmarks for trace collinearity and its relationship to best fit plane quality to aid the practical implementation of the results of this study. Furthermore, we supplement quantitative comparative analysis with recommended best practices for 3D trace digitization, such as ensuring high values of coplanarity, maintaining a quasi-constant node-to-node distance relative to the model's resolution, and ensuring a minimum number of nodes to guarantee the robustness of the fitted planar model. Critically, our study highlights the critical role tacit geological knowledge plays in the robustness of 3D trace digitization from virtual outcrop models.

Farm Plot Boundary Estimation and Testing Based on the Digital Filtering and Integral Clustering of Seeding Trajectories

Farmland boundary data, an important basic data for the operation of agricultural automation equipment, has been widely studied by scholars from all over the world. However, the common methods of farmland boundary acquisition through sensors such as LiDAR and vision cameras combined with complex algorithms suffer from problems such as serious data drift, difficulty in eliminating noise, and inaccurate plot boundary data. In order to solve this problem, this study proposes a method for estimating the orientation dimensions of farmland based on the seeding trajectory. The method firstly calculates the curvature of the discrete data of the seeding trajectory; secondly, we innovatively use a low-pass filter and integral clustering to filter the curvature values and distinguish between straight lines and curves; and finally, the straight-line portion located at the edge of the seeding trajectory is fitted with a univariate linear fit to calculate the estimation of the farmland size orientation. As verified by the field experiments, the minimum linear error of the vertices is only 0.12m, the average error is 0.315m, and the overlapping rate of the plot estimation is 98.36% compared with the real boundary of the plot. Compared with LiDAR mapping, the average linear error of the vertices’ position is reduced by 50.2%, and the plot estimation overlap rate is increased by 2.21%. The experimental results show that this method has the advantage of high accuracy, fast calculation speed, and small calculation volume, which provides a simple and accurate method for constructing farmland maps, provides the digital data support for the operation of agricultural automation equipment, and has significance for farm digital mapping.

Interference across time: dissociating short from long temporal interference.

Our ability to identify an object is often impaired by the presence of preceding and/or succeeding task-irrelevant items. Understanding this temporal interference is critical for any theoretical account of interference across time and for minimizing its detrimental effects. Therefore, we used the same sequences of 3 orientation items, orientation estimation task, and computational models, to examine temporal interference over both short (<150 ms; visual masking) and long (175-475 ms; temporal crowding) intervals. We further examined how inter-item similarity modifies these different instances of temporal interference. Qualitatively different results emerged for interference of different scales. Interference over long intervals mainly degraded the precision of the target encoding while interference over short intervals mainly affected the signal-to-noise ratio. Although both interference instances modulated substitution errors (reporting a wrong item) and were alleviated with dissimilar items, their characteristics were markedly disparate. These findings suggest that different mechanisms mediate temporal interference of different scales.

Signals-of-Opportunity-Based Hydrophone Array Shape and Orientation Estimation

Signals-of-Opportunity-Based Hydrophone Array Shape and Orientation Estimation

Empowering lightweight detectors: Orientation Distillation via anti-ambiguous spatial transformation for remote sensing images

Knowledge distillation (KD) has been one of the most potential methods to implement a lightweight detector, which plays a significant role in satellite in-orbit processing and unmanned aerial vehicle tracking. However, existing distillation paradigms exhibit limited accuracy in detecting arbitrary-oriented objects represented with rotated bounding boxes in remote sensing images. This issue is attributed to two aspects: (i) boundary discontinuity localization distillation, caused by angle periodicity of rotated bounding boxes, and (ii) spatial ossified feature distillation, induced by orientation-agnostic knowledge transitive regions, both of which contribute to ambiguous orientation estimation of objects. To address these issues, we propose an effective KD method called Orientation Distillation (OD) via anti-ambiguous spatial transformation, which consists of two modules. (i) Anti-ambiguous Location Prediction (ALP) module reformulates the regression transformation between teacher–student bounding boxes as Gaussian distributions fitting procedure. These distributions with distilled potential are optimized to accurately localize objects with the aid of boundary continuity cost. (ii) Orientation-guided Feature Calibration (OFC) module employs a learnable affine matrix to augment fixed CNN sampling grid into a spatially remapped one, which bridges between the multi-scale feature of teacher and student for effectively delivering the refined oriented awareness within adaptively distillation regions. Overall, OD customizes the spatial transformation of bounding box representation and sampling grid to transfer anti-ambiguous orientation knowledge, and significantly improves the performance of lightweight detectors upon non-axially arranged objects. Extensive experiments on multiple datasets demonstrate that our plug-and-play distillation framework achieves state-of-the-art performance. Codes are available at https://github.com/Molly6/OD.

Revealing asymmetry on mid-plane of protoplanetary disc through modelling of axisymmetric emission: methodology

ABSTRACT This study proposes an analytical framework for deriving the surface brightness profile and geometry of a geometrically thin axisymmetric disc from interferometric observation of continuum emission. Such precise modelling facilitates the exploration of faint non-axisymmetric structures, such as spirals and circumplanetary discs. As a demonstration, we simulate interferometric observations of geometrically thin axisymmetric discs. The proposed method can reasonably recover the injected axisymmetric structures, whereas Gaussian fitting of the same data yielded larger errors in disc orientation estimation. To further test the applicability of the method, it was applied to the mock data for $m=1,2$ spirals and a point source, which are embedded in a bright axisymmetric structure. The injected non-axisymmetric structures were reasonably recovered except for the innermost parts, and the disc geometric parameter estimations were better than Gasussian fitting. The method was then applied to the real data of Elias 20 and AS 209, and it adequately subtracted the axisymmetric component, notably in Elias 20, where substantial residuals remained without our method. We also applied our method to continuum data of PDS 70 to demonstrate the effectiveness of the method. We successfully recovered emission from PDS 70 c consistently with previous studies, and also tentatively discovered new substructures. The current formulation can be applied to any data for disc continuum emission, and aids in the search of spirals and circumplanetary discs, whose detection is still limited.

Joint stereo 3D object detection and implicit surface reconstruction

We present a new learning-based framework S-3D-RCNN that can recover accurate object orientation in SO(3) and simultaneously predict implicit rigid shapes from stereo RGB images. For orientation estimation, in contrast to previous studies that map local appearance to observation angles, we propose a progressive approach by extracting meaningful Intermediate Geometrical Representations (IGRs). This approach features a deep model that transforms perceived intensities from one or two views to object part coordinates to achieve direct egocentric object orientation estimation in the camera coordinate system. To further achieve finer description inside 3D bounding boxes, we investigate the implicit shape estimation problem from stereo images. We model visible object surfaces by designing a point-based representation, augmenting IGRs to explicitly address the unseen surface hallucination problem. Extensive experiments validate the effectiveness of the proposed IGRs, and S-3D-RCNN achieves superior 3D scene understanding performance. We also designed new metrics on the KITTI benchmark for our evaluation of implicit shape estimation.

Deep learning model for automatic limb detection and gripping in a novel meat factory cell

The goal of this work has been the development of a deep-learning model which may be used in conjunction with a novel so-called “Meat Factory Cell” platform, namely to enable an industrial robot within the system to identify and successfully grip the limbs of an entire pig carcass. The model consists of three main components: (1) a U-Net-based deep learning model that predicts heatmaps with a probability distribution of gripping and key point locations on the limbs, within RGB-D images; (2) a post-processing element for the extraction of keypoints from heatmaps, and transferral of these points into 3D space using a pinhole camera model; and (3) gripper orientation estimation, which uses the predicted limb key points to define gripper orientation in 3D space. The proposed system demonstrates high precision and robustness in estimating gripping points on pig limbs based on a data test set, which includes two gripping definitions: Norwegian and Danish. These gripping definitions account for variation in the slaughter process in two different European countries. The Norwegian definition gives mAP(0.5…0.95) = 0.971, mAR(0.5…0.95) = 0.982, and distance error 13 mm, while the Danish definition gives mAP(0.5…0.95) = 0.985, mAR(0.5…0.95) = 0.995, and distance error 14 mm. The model was validated in practice during experimental trials at the Meat Factory Cell test facility at the Norwegian University of Life Science (Ås, Norway), with whole pig carcasses (n=25).

Construction Activity Recognition Method Based on Object Detection, Attention Orientation Estimation, and Person Re-Identification

Recognition and classification for construction activities help to monitor and manage construction workers. Deep learning and computer vision technologies have addressed many limitations of traditional manual methods in complex construction environments. However, distinguishing different workers and establishing a clear recognition logic remain challenging. To address these issues, we propose a novel construction activity recognition method that integrates multiple deep learning algorithms. To complete this research, we created three datasets: 727 images for construction entities, 2546 for posture and orientation estimation, and 5455 for worker re-identification. First, a YOLO v5-based model is trained for worker posture and orientation detection. A person re-identification algorithm is then introduced to distinguish workers by tracking their coordinates, body and head orientations, and postures over time, then estimating their attention direction. Additionally, a YOLO v5-based object detection model is developed to identify ten common construction entity objects. The worker’s activity is determined by combining their attentional orientation, positional information, and interaction with detected construction entities. Ten video clips are selected for testing, and a total of 745 instances of workers are detected, achieving an accuracy rate of 88.5%. With further refinement, this method shows promise for a broader application in construction activity recognition, enhancing site management efficiency.

A Novel IMU-Based System for Work-Related Musculoskeletal Disorders Risk Assessment.

This study introduces a novel wearable Inertial Measurement Unit (IMU)-based system for an objective and comprehensive assessment of Work-Related Musculoskeletal Disorders (WMSDs), thus enhancing workplace safety. The system integrates wearable technology with a user-friendly interface, providing magnetometer-free orientation estimation, joint angle measurements, and WMSDs risk evaluation. Tested in a cable manufacturing facility, the system was evaluated with ten female employees. The evaluation involved work cycle identification, inter-subject comparisons, and benchmarking against standard WMSD risk assessments like RULA, REBA, Strain Index, and Rodgers Muscle Fatigue Analysis. The evaluation demonstrated uniform joint patterns across participants (ICC=0.72±0.23) and revealed a higher occurrence of postures warranting further investigation, which is not easily detected by traditional methods such as RULA. The experimental results showed that the proposed system's risk assessments closely aligned with the established methods and enabled detailed and targeted risk assessments, pinpointing specific bodily areas for immediate ergonomic interventions. This approach not only enhances the detection of ergonomic risks but also supports the development of personalized intervention strategies, addressing common workplace issues such as tendinitis, low back pain, and carpal tunnel syndrome. The outcomes highlight the system's sensitivity and specificity in identifying ergonomic hazards. Future efforts should focus on broader validation and exploring the relative influence of various WMSDs risk factors to refine risk assessment and intervention strategies for improved applicability in occupational health.

Convolutional Cross-View Pose Estimation.

We propose a novel end-to-end method for cross-view pose estimation. Given a ground-level query image and an aerial image that covers the query's local neighborhood, the 3 Degrees-of-Freedom camera pose of the query is estimated by matching its image descriptor to descriptors of local regions within the aerial image. The orientation-aware descriptors are obtained by using a translationally equivariant convolutional ground image encoder and contrastive learning. The Localization Decoder produces a dense probability distribution in a coarse-to-fine manner with a novel Localization Matching Upsampling module. A smaller Orientation Decoder produces a vector field to condition the orientation estimate on the localization. Our method is validated on the VIGOR and KITTI datasets, where it surpasses the state-of-the-art baseline by 72% and 36% in median localization error for comparable orientation estimation accuracy. The predicted probability distribution can represent localization ambiguity, and enables rejecting possible erroneous predictions. Without re-training, the model can infer on ground images with different field of views and utilize orientation priors if available. On the Oxford RobotCar dataset, our method can reliably estimate the ego-vehicle's pose over time, achieving a median localization error under 1 m and a median orientation error of around 1 ° at 14 FPS.

A photogrammetric approach for real‐time visual SLAM applied to an omnidirectional system

AbstractThe problem of sequential estimation of the exterior orientation of imaging sensors and the three‐dimensional environment reconstruction in real time is commonly known as visual simultaneous localisation and mapping (vSLAM). Omnidirectional optical sensors have been increasingly used in vSLAM solutions, mainly for providing a wider view of the scene, allowing the extraction of more features. However, dealing with unmodelled points in the hyperhemispherical field poses challenges, mainly due to the complex lens geometry entailed in the image formation process. To address these challenges, the use of rigorous photogrammetric models that appropriately handle the geometry of fisheye lens cameras can overcome these challenges. Thus, this study presents a real‐time vSLAM approach for omnidirectional systems adapting ORB‐SLAM with a rigorous projection model (equisolid‐angle). The implementation was conducted on the Nvidia Jetson TX2 board, and the approach was evaluated using hyperhemispherical images captured by a dual‐fisheye camera (Ricoh Theta S) embedded into a mobile backpack platform. The trajectory covered a distance of 140 m, with the approach demonstrating accuracy better than 0.12 m at the beginning and achieving metre‐level accuracy at the end of the trajectory. Additionally, we compared the performance of our proposed approach with a generic model for fisheye lens cameras.

Direct orientation estimation through inertial odometry based on a deep transformer model

Direct orientation estimation through inertial odometry based on a deep transformer model

Enhancing accuracy and convenience of golf swing tracking with a wrist-worn single inertial sensor

In this study, we address two technical challenges to enhance golf swing trajectory accuracy using a wrist-worn inertial sensor: orientation estimation and drift error mitigation. We extrapolated consistent sensor orientation from specific address-phase signal segments and trained the estimation with a convolutional neural network. We then mitigated drift error by applying a constraint on wrist speed at the address, backswing top, and finish, and ensuring that the wrist's finish displacement aligns with a virtual circle on the 3D swing plane. To verify the proposed methods, we gathered data from twenty male right-handed golfers, including professionals and amateurs, using a driver and a 7-iron. The orientation estimation error was about 60% of the baseline, comparable to studies requiring additional sensor information or calibration poses. The drift error was halved and the single-inertial-sensor tracking performance across all swing phases was about 17 cm, on par with multimodal approaches. This study introduces a novel signal processing method for tracking rapid, wide-ranging motions, such as a golf swing, while maintaining user convenience. Our results could impact the burgeoning field of daily motion monitoring for health care, especially with the increasing prevalence of wearable devices like smartwatches.

Robust Orientation Estimation from MEMS Magnetic, Angular Rate, and Gravity (MARG) Modules for Human-Computer Interaction.

While the availability of low-cost micro electro-mechanical systems (MEMS) accelerometers, gyroscopes, and magnetometers initially seemed to promise the possibility of using them to easily track the position and orientation of virtually any object that they could be attached to, this promise has not yet been fulfilled. Navigation-grade accelerometers and gyroscopes have long been the basis for tracking ships and aircraft, but the signals from low-cost MEMS accelerometers and gyroscopes are still orders of magnitude poorer in quality (e.g., bias stability). Therefore, the applications of MEMS inertial measurement units (IMUs), containing tri-axial accelerometers and gyroscopes, are currently not as extensive as they were expected to be. Even the addition of MEMS tri-axial magnetometers, to conform magnetic, angular rate, and gravity (MARG) sensor modules, has not fully overcome the challenges involved in using these modules for long-term orientation estimation, which would be of great benefit for the tracking of human-computer hand-held controllers or tracking of Internet-Of-Things (IoT) devices. Here, we present an algorithm, GMVDμK (or simply GMVDK), that aims at taking full advantage of all the signals available from a MARG module to robustly estimate its orientation, while preventing damaging overcorrections, within the context of a human-computer interaction application. Through experimental comparison, we show that GMVDK is more robust to magnetic disturbances than three other MARG orientation estimation algorithms in representative trials.

Pre-rotation Only at Inference-Stage: A Way to Rotation Invariance of Convolutional Neural Networks

The popular convolutional neural networks (CNN) require data augmentation to achieve rotation invariance. We propose an alternative mechanism, Pre-Rotation Only at Inference stage (PROAI), to make CNN rotation invariant. The overall idea is to learn how the human brain observe images. At the training stage, PROAI trains a CNN with a small number using images only at one orientation. At the inference stage, PROAI introduces a pre-rotation operation to rotate each test image into its all-possible orientations and calculate classification scores using the trained CNN with a small number of parameters. The maximum of these classification scores is able to simultaneously estimate both the category and the orientation of each test image. The specific benefits of PROAI have been experimented on rotated image recognition tasks. The results shows that PROAI improves both the classification and orientation estimation performance while greatly reduced the numbers of parameters and the training time. Codes and datasets are publicly available at https://github.com/automlresearch/FRPRF.

IMU-Assisted Manual 3D-Ultrasound Imaging Using Motion-Constrained Swept-Fan Scans.

Three-dimensional (3D) ultrasonic imaging can enable post-facto plane of interest selection. It can be performed with devices such as wobbler probes, matrix probes, and sensor-based probes. Ultrasound systems that support 3D-imaging are expensive with added hardware complexity compared to 2D-imaging systems. An inertial measurement unit (IMU) can potentially be used for 3D-imaging by using it to track the motion of a one-dimensional array probe and constraining its motion in one degree of freedom (1-DoF) rotation (swept-fan). This work demonstrates the feasibility of an affordable IMU-assisted manual 3D-ultrasound scanner (IAM3US). A consumer-grade IMU-assisted 3D scanner prototype is designed with two support structures for swept-fan. After proper IMU calibration, an appropriate KF-based algorithm estimates the probe orientation during the swept-fan. An improved scanline-based reconstruction method is used for volume reconstruction. The evaluation of the IAM3US system is done by imaging a tennis ball filled with water and the head region of a fetal phantom. From fetal phantom reconstructed volumes, suitable 2D planes are extracted for biparietal diameter (BPD) manual measurements. Later, in-vivo data is collected. The novel contributions of this paper are (1) the application of a recently proposed algorithm for orientation estimation of swept-fan for 3D imaging, chosen based on the noise characteristics of selected consumer grade IMU (2) assessment of the quality of the 1-DoF swept-fan scan with a deflection detector along with monitoring of maximum angular rate during the scan and (3) two probe holder designs to aid the operator in performing the 1-DoF rotational motion and (4) end-to-end 3D-imaging system-integration. Phantom studies and preliminary in-vivo obstetric scans performed on two patients illustrate the usability of the system for diagnosis purposes.

Angle-corrected GPR hyperbolic fitting models for improved parameter estimation

Ground-penetrating radar (GPR) is commonly employed as a non-destructive technique for detecting subsurface cylindrical objects such as pipes, cables, rebars, and tree roots. Hyperbolic features generated in GPR radargrams are often used to estimate key parameters like burial depth, object radius, and electromagnetic wave velocity. However, traditional approaches frequently rely on the assumption that GPR traverses are perpendicular to the alignment of target—an assumption or a negligence that is not always valid in real-world scenarios. To address this limitation, a novel method is introduced for simultaneously estimating the orientation and burial depth of pipes, as well as wave velocity, from the hyperbolic patterns observed in GPR data. The method innovates by incorporating an angle correction index into the classical hyperbolic fitting model. This modified model is then formulated as an optimization problem, which is solved using a hybrid approach combining the Multi-Verse Optimizer (MVO) and Gradient Descent (GD) algorithms. A unique index, termed the “C-value,” is introduced to quantitatively analyse the influence of oblique angles on the hyperbolic fitting models. Two distinct fitting models are validated through both simulation and field experiments. The study also scrutinizes the impact of varying pipe radius and burial depth on the accuracy of parameter estimation at different pipe orientation. The methodology presented herein enables the simultaneous estimation of burial depth, wave velocity, and pipe orientation directly from hyperbolic fitting—a significant advancement, as orientation is typically assumed as a known input yet is often challenging to ascertain obtain in practical field situations.

Temporal integration of target features across and within trials in the attentional blink.

Attentional blink research has typically investigated attentional limitations in multiple target processing. The current study investigated the temporal integration of target features in the attentional blink. Across two experiments, we demonstrated that the orientation estimations of individual target items in the attentional blink paradigm were systematically biased. Specifically, there was evidence for both within- and across-trial biases, revealing a general bias towards previously presented stimuli. Moreover, both biases were found to be more salient for targets suffering from the attentional blink. The current study is the first to demonstrate an across-trial bias in responses in the attentional blink paradigm. This set of findings is in line with the literature, suggesting that the human visual system can implicitly summarize information presented over time, which may lead to biases. By investigating temporal integration in the attentional blink, we have been able to address the modulatory role of attention on biases imposed by the implicit temporal effects in estimation tasks. Our findings may inform future research on attentional blink, serial dependence, and ensemble perception.

Analysis of the Influence of Polarization Measurement Errors on the Parameter and Characteristics Measurement of the Fully Polarized Entomological Radar

Measuring the orientation, mass and body length of migratory insects through entomological radar is crucial for early warnings of migratory pests. The fully polarized entomological radar is an efficient device for observing migratory insects by calculating insect parameters through the scattering matrix obtained from the target. However, the measured target scattering matrix will be affected by system polarization measurement errors, leading to errors in insect parameter calculation, while the related analysis is currently relatively limited. Therefore, the scattering matrix measurement process is modeled, followed by an analysis of the effects of different errors on orientation, mass and body length estimation. The influence of polarization measurement errors on insect scattering characteristics is also analyzed. The results present that for fixed polarization measurement errors, the measurement errors of insect orientation, mass and body length will vary with insect orientation in specific patterns, and the distribution of measured insect parameters will be drastically distorted compared to the true parameter distribution. In addition, polarization measurement errors could seriously disrupt the reciprocity and bilateral symmetry of the measured insect scattering matrix. These analyses and conclusions provide a good basis for eliminating the effects of polarization measurement errors and improving the accuracy of insect parameter measurement.