Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
42
datasets available to search
ShareScore release 0.7.1
Dataset results
42 results for “3D tracking”
Feasibility of 3D Body Tracking from Monocular 2D Video Feeds in Musculoskeletal Telerehabilitation
<p>Musculoskeletal conditions affect millions of people globally, however, conventional treatments pose challenges concerning price, accessibility, and convenience. Many telerehabilitation solutions offer an engaging alternative but rely on complex hardware for body tracking. This work explores the feasibility of models for 3D Human Pose Estimation (HPE) from monocular 2D videos (MediaPipe Pose) in a physiotherapy context, by comparing its performance to ground truth measurements. MediaPipe Pose was investigated in eight exercises typically performed in musculoskeletal physiotherapy sessions, where the Range of Motion (ROM) of the human joints was the evaluated parameter. This model showed the best performance for shoulder abduction, shoulder press, elbow flexion, and squat exercises (MAPE ranging between 14.9% and 25.0%, Pearson’s coefficient ranging between 0.963 and 0.996, and cosine similarity ranging between 0.987 and 0.999). Some exercises (e.g. seated knee extension and shoulder flexion) posed challenges due to unusual poses, occlusions and depth ambiguities, possibly related to a lack of training data. This study demonstrates the potential of HPE from monocular 2D videos, as a markerless, affordable and accessible solution for musculoskeletal telerehabilitation approaches. Future work should focus on exploring variations of the 3D HPE models trained on physiotherapy-related datasets, such as the Fit3D dataset, and post-preprocessing techniques to enhance the model's performance.</p>
Next-generation 3D object detection and tracking for self-driving vehicles using object velocity
<p>The synthetic dataset was generated using KITTI-like specifications and annotations format. It is comprised by the training and testing sets, that include KITTI standard folders: label_2, image_2 and calib. Furthermore, there is a velodyne file for each of the following use cases:</p><ul><li>Point cloud 1: (x,y,z, (Float)Radial_Velocity): this point cloud has the relative radial velocity as an additional feature for each point. File: velodyne_radial_velocity;</li><li>Point cloud 2: (x,y,z,(Float)Absolute_Speed): in this point cloud, every point has the absolute speed of the object as the additional feature. File: velodyne_abs_speed;</li><li>Point cloud 3: (x,y,z,(Bool)Is_Moving): the additional feature of this point cloud is a Boolean value that is set to 1.0 if the object is moving; contrariwise, it is set to 0.0 for static objects. File: velodyne_is_moving;</li><li>Point cloud 4: (x,y,z,0): no additional feature information. If desired, requires post-processing to convert to (x,y,z) or changing the toolbox point cloud configuration to not consider the additional feature. File: velodyne_xyz;</li></ul><p>Additionally, the detections generated with the OpenPCDet toolbox and Second-IoU model are provided.</p><p>This work was made as part of a master thesis of Informatics Engineering in the University of Aveiro.</p>
3D cell tracking dataset of bacterial biofilm deformation and recovery under shear flow
<p>This MAT file includes dataset in the scientific article "<i>In vivo</i> microrheology reveals elastic and plastic responses inside three-dimensional bacterial biofilms" by the following authors: Takuya Ohmura, Dominic Skinner, Konstantin Neuhaus, Gary Choi, Jörn Dunkel, Knut Drescher. This MAT file can be conveniently opened with Matlab. </p><p>When you open this file with Matlab, you will find 4 variables stored in the file "Data_v3_loop2_newRxy_bidx1_274.mat"</p><p><strong>Variable 1: name_parameter</strong></p><p>Names of 31 parameters for columns in 3 variables: 'deformation_all', 'recovery_all' , 'plasticity_all'. The parameters have cell displacements, orientations, coordinates, biofilm indexes and experimental conditions. When the parameters have units, they are shown in the names. </p><ul><li>'x_Frame1[um]'</li><li>'y_Frame1[um]'</li><li>'z_Frame1[um]'</li><li>'Normalized_x_Frame1'</li><li>'Normalized_y_Frame1'</li><li>'Normalized_z_Frame1'</li><li>'LocalDensity_Frame1(VolumeFractionAround30px)'</li><li>'LocalCellNumberDensity_Frame1(VolumeFractionAround30px)'</li><li>'NematicOrderParameter_Frame1'</li><li>'AlignmentFlow_Frame1[rad]'</li><li>'AlignmentRadial_Frame1[rad]'</li><li>'AlignmentZaxis_Frame1[rad]'</li><li>'d_x[um]'</li><li>'d_y[um]'</li><li>'d_z[um]'</li><li>'Normalized_d_x'</li><li>'Normalized_d_y'</li><li>'Normalized_d_z'</li><li>'d_LocalDensity'</li><li>'d_LocalNumberDensity[um^-3]'</li><li>'d_NematicOrderParameter'</li><li>'d_AlignmentFlow[rad]'</li><li>'d_AlignmentRadial[rad]'</li><li>'d_AlignmentZaxis[rad]'</li><li>'CrossProduct[um^2]'</li><li>'BiofilmIndexNumber'</li><li>'Biofilm_width[um]'</li><li>'Biofilm_height[um]'</li><li>'Biofilm_volume[um^3]'</li><li>'FlowRate[ul/min]'</li><li>'Duration[min]'</li></ul><p><strong>Variable 2: deformation_all</strong></p><p>The rows indicate 704198 single-cell trackings in deformations of 274 bacterial biofilms. Each of the 274 bacterial biofilms has a different 'BiofilmIndexNumber'. The columns indicate 31 parameters which names are shown in 'name_parameter'.</p><p><strong>Variable 3: recovery_all</strong></p><p>The rows indicate 685991 single-cell trackings in recoveries of 274 bacterial biofilms. Each of the 274 bacterial biofilms has a different 'BiofilmIndexNumber'. The columns indicate 31 parameters which names are shown in 'name_parameter'.</p><p><strong>Variable 4: plasticity_all</strong></p><p>The rows indicate 665749 single-cell trackings in plasticities of 274 bacterial biofilms. Each of the 274 bacterial biofilms has a different 'BiofilmIndexNumber'. The columns indicate 31 parameters which names are shown in 'name_parameter'.</p><p> </p><p>To plot the cell tracked data in the figures of the article, use our MATLAB code uploaded in our GitHub (https://github.com/knutdrescher/biofilm-rheology).</p>
FIGURE 8. Molfetta dinosaur tracks, photogrammetry derived 3D in The use of aerial and close-range photogrammetry in the study of dinosaur tracksites: Lower Cretaceous (upper Aptian/lower Albian) Molfetta ichnosite (Apulia, southern Italy)
FIGURE 8. Molfetta dinosaur tracks, photogrammetry derived 3D models and interpretations; 1-2, DEM and contour line map of a theropod footprint (contour lines have an interval of 0.2 cm); 4-5, DEM and contour line map of an ornitischian footprint; 3 and 6, interpretative outline drawings of the studied tracks.
FIGURE 6. Shaded coloured 3D in Elongated theropod tracks from the Cretaceous Apenninic Carbonate Platform of southern Latium (central Italy
FIGURE 6. Shaded coloured 3D photogrammetric model of footprint F1 and F2 with relative section (S4) passing through the metatarsal impression and digit III.
FIGURE 5. Shaded coloured 3D in Elongated theropod tracks from the Cretaceous Apenninic Carbonate Platform of southern Latium (central Italy
FIGURE 5. Shaded coloured 3D photogrammetric model of footprint F3 and relative sections (S1, S2, S3).
FIGURE 4. 1, Shaded grey 3D in Elongated theropod tracks from the Cretaceous Apenninic Carbonate Platform of southern Latium (central Italy
FIGURE 4. 1, Shaded grey 3D photogrammetric model of the track-bearing block; 2, Shaded coloured 3D photogrammetric model.
Outputs from new methods for 3D+time cell image segmentation and tracking
<p>Segmentation and tracking of 3D+time microscopy images of cell nuclei within the zebrafish pectoral fin.</p> <p>The file named 7_cells_moving_in_70_frames_orig.avi is a 70-frame video of a group of cells moving in time, the file named 7_cells_moving_in_70_frames.avi has the result of 4D segmentation, using our new segmentation methods, for seven cells (colored black) moving in time, and the file _tracking_of_7_cell_in_70_frames.mp4 has the tracking of these seven cells. </p> <p>Additionally, the file named group_of_cells_moving_in_70_frames_orig.avi is a 70-frame video of a group of cells moving in time, the file named group_of_cells_moving_in_70_frames.avi has the result of 4D segmentation, using our new segmentation methods, for the group of cells (colored black) moving in time and the file _tracking_of_group_of_cell_in_70_frames.gif has the tracking of this group of cells. </p>
Imaging data from "Live-cell 3D single-molecule tracking reveals modulation of enhancer dynamics by NuRD"
<p>3D 20ms, 3D 500ms and 2D dCas9 raw videos, localisation, tracking and trajectory analysis data</p> <p>From 'Live-cell 3D single-molecule tracking reveals modulation of enhancer dynamics by NuRD" (2021). Biorxiv. https://doi.org/10.1101/2020.04.03.003178</p>
Data from: Long-term monitoring of <em>Ziphius cavirostris</em> behavior using 3D tracking from fixed hydrophone arrays off Southern California
Open the record for dataset details and reuse information.
Audio and 3D flight-track recordings of mosquito responses to opposite-sex sound-stimuli
Open the record for dataset details and reuse information.
3D Matching of resource vision tracking trajectories
<p>Three dimensional (3D) paths of resources, have been proposed in construction management, as an efficient way for measuring labor productivity. These paths, are extracted either by using sensors such as Global Positioning System (GPS), Radio Frequency Identification (RFID), and Ultra-wideband (UWB), or based on cameras placed at jobsites for surveillance purposes. However, the tag based methods are seriously limited by privacy conflicts since they are not welcome from the personnel. On the other hand, the computer vision based methods have not achieved full automation in measuring labour productivity because they require prior knowledge of the type of tasks performed in specific working zones. This is associated with the lack of depth information. For this purpose, this paper proposes a computationally efficient computer vision method for matching construction workers across different frames. Entity matching, is a process that corresponds to a compulsory step prior to the calculation of the 3D position. The proposed matching method, is based on epipolar geometry, template and motion similarity features. The main result of this process, is to provide a method for the acquisition of the 3D paths that compose the detailed profile of a construction activity in terms of both time and space.</p>
FISH datasets used in Zou et al. integrating multi-track Hi-C data for genome-scale reconstruction of 3D chromatin structure
<p>This upload contains the FISH datasets used in Zou et al. integrating multi-track Hi-C data for genome-scale reconstruction of 3D chromatin structure.</p> <p>If you use the datasets, we would be grateful if you cited the following paper:</p> <p>Zou, C., Zhang, Y., Ouyang, Z. (2016) HSA: integrating multi-track Hi-C data for genome-scale reconstruction of 3D chromatin structure. Genome Biology, 17: 40.</p>
Matching Construction Workers across Views for Automated 3D Vision Tracking On-Site
<p>Computer vision based tracking of construction resources is one of the several options available for obtaining trajectories useful in safety and productivity applications. This type of tracking requires that targets are accurately matched across multiple camera views to obtain a 3D trajectory out of two or more 2D trajectories. This matching is straightforward when it involves easily distinguishable targets in uncluttered scenes. This can be challenging in industrial scenes such as construction sites due to congestion, occlusions and workers in greatly similar high visibility apparel. This paper proposes a novel vision based method that addresses all these issues. It uses as input the output of a 2D vision based tracking method and searches for potential matches in three sequential steps. It terminates only when a positive match is found. The first step returns the strongest candidate by correlating a segment of workers’ past 2D trajectories. The second employs geometric restrictions, whilst the third correlates colour intensity values. The proposed method featured a promising performance of 97% precision, 98% recall and 95% accuracy.</p>
HT1080WT cells embedded in 3D collagen type I matrices - manual annotations for cell instance segmentation and tracking
<p>Human fibrosarcoma HT1080WT (ATCC) cells at low cell densities embedded in 3D collagen type I matrices [1]. The time-lapse videos were recorded every 2 minutes for 16.7 hours and covered a field of view of 1002 pixels × 1004 pixels with a pixel size of 0.802 μm/pixel The videos were pre-processed to correct frame-to-frame drift artifacts, resulting in a final size of 983 pixels × 985 pixels pixels.</p> <p><em>Hasini Jayatilaka, Anjil Giri, Michelle Karl, Ivie Aifuwa, Nicholaus J Trenton, Jude M Phillip, Shyam Khatau, and Denis Wirtz. EB1 and cytoplasmic dynein mediate protrusion dynamics for efficient 3-dimensional cell migration. FASEB J., 32(3):1207–1221, 2018. ISSN 0892-6638. doi: 10.1096/fj.201700444RR.</em></p> <p>Further information about how to use this data is given in <a href="http://github.com/esgomezm/microscopy-dl-suite-tf">https://github.com/esgomezm/microscopy-dl-suite-tf</a></p> <p><strong>This dataset is provided together with the following preprint and if you use it, we would like to kindly ask you to cite it properly:</strong></p> <p><a href="https://arxiv.org/abs/2112.08817">Estibaliz Gómez-de-Mariscal, Hasini Jayatilaka, Özgün Çiçek, Thomas Brox, Denis Wirtz, Arrate Muñoz-Barrutia, *Search for temporal cell segmentation robustness in phase-contrast microscopy videos*, arXiv 2021 (arXiv:2112.08817)</a></p>
Particle tracks of 3D flow field during the initiation of a fluvial particle (case 6S)
<p>The dataset belongs to experimental work described in the paper "Role of low-order proper orthogonal decomposition modes and large-scale coherent structures on sediment particle entrainment" published in the Journal of Hydraulics Research. The paper has been published as open access: https://doi.org/10.1080/00221686.2020.1869604</p> <p>The dataset consists of particle tracks of the three-dimensional flow field during the entrainment of a fluvial particle of test case 6S.</p> <p> </p> <p> </p>
3D+time nuclei tracking dataset of confocal fluorescence microscopy time series of C. elegans embryos
<p>The dataset consists of 3 confocal microscopy time series of <em>C. elegans</em> embryos, fully tracked with StarryNite followed by manual curation</p> <ul> <li>3 raw time-series and the corresponding tracks/lineage trees</li> <li>temporal resolution; 75s</li> <li>temporal extent: 400 frames, tracked for at least 370 frames</li> <li>spatial resolution (zyx): 0.75 x 0.15 x 0.15 μm</li> <li>spatial extent (zyx):/ 41 x 512 x 512px</li> <li>Microscope: Zeiss Axio Observer.Z1</li> </ul> <p>The annotations were created using the method described in:</p> <p><em> Santella, A., Du, Z. & Bao, Z. A semi-local neighborhood-based framework for probabilistic cell lineage tracing. BMC Bioinformatics 15, 217 (2014). <a href="https://doi.org/10.1186/1471-2105-15-217">https://doi.org/10.1186/1471-2105-15-217</a></em></p> <p>Additionally the data was extended and used for the development of a new tracking method in the following publication:</p> <p><em> Hirsch, P., Malin-Mayor C., Santella, A., Preibisch, S., Kainmueller, D., Funke, J. Tracking by weakly-supervised learning and graph optimization for whole-embryo C. elegans lineages. MICCAI 2022.</em></p> <p>For questions please contact Peter Hirsch (<a href="mailto:peter.hirsch@mdc-berlin.de">peterhirsch@posteo.de</a>).</p>
Data: UTrack3D: 3D Tracking Using Ultra-wideband (UWB) Radios
<p>"# UTrack3D"</p> <p><strong>Environments</strong>: Python3.7 & Matlab2021</p> <p><strong>System</strong>: Windows 11</p> <div> <h2>Install prerequisites</h2> <a href="https://github.com/yifeng361/UTrack3D#install-prerequisites"></a></div> <ul> <li> <p>Python: We test our code using Python3.7. Advanced Python versions work as well. <a href="https://www.python.org/downloads/" rel="nofollow">https://www.python.org/downloads/</a></p> </li> <li> <p>Matlab: We test our code using Matlab2021. Advanced Matlab versions work as well. The matlab is only used for performance evaluation in this code. In case you prefer not installing Matlab, several pre-generated examples are provided.</p> </li> <li> <p>Python libraries: <code>pip install -r requirements.txt</code></p> </li> </ul> <div> <h2>Running</h2> <a href="https://github.com/yifeng361/UTrack3D#running"></a></div> <ul> <li>Run script <code>run_offline_analysis.py</code> for tracking.</li> </ul> <p><code>python run_offline_track.py</code></p> <p>This reads pre-stored CIR data (./raw_data) and generates a file <code>tracking_results.mat</code> in ./output which stores the estimated trajectory and ground-truth trajectory. We provide three examples (test1, test2, test3). One can modify the following line to test a specific example.</p> <p><code>file_dir = "./raw_data/test1/"</code></p> <ul> <li>Run script <code>./matlab_analysis_scripts/evaluate_accuracy_ae.m</code> to compute error and perform visualization. This script takes <code>tracking_results.mat</code> as inputs and generates CDF error plot and trajectory visualization in the current folder.</li> </ul> <p>The CDF error plot and trajectories of three examples have already been pre-generated and put in <code>./matlab_analysis_scripts/</code>.</p>
Dataset for Neural Network 3D Body Pose Tracking and Prediction for Motion-to-Photon Latency Compensation in Distributed Virtual Reality
<p>Distributed Virtual Reality (DVR) systems enable geographically dispersed users to interact in a shared virtual environment. The realism of the interaction is crucial to increase the feeling of co-presence. Latency, produced either by hard- or software components of DVR applications, impedes reaching high realism levels of the DVR experience. For example, the time delay between the user's motion and the corresponding display rendering of the DVR system might lead to adverse effects such as a reduced sense of presence or motion sickness. One way of minimizing the latency is to predict user's motion and thus compensate for the inherent latency in the system. In order to address this problem, we propose a neural network 3D pose tracking and prediction system with latency guarantees for end-to-end avatar reconstruction. We evaluate and compare our system against multiple traditional methods and provide a thorough analysis on real-world human motion data. Datasets used in the paper experiments. Datasets used in paper experiments.</p>
Stereo video files used for 3d tracking horsefly trajectories
<p>Of all hypotheses advanced for why zebras have stripes, avoidance of biting fly attack receives by far the most support, yet the mechanisms by which stripes thwart landings are not yet understood. A logical and popular hypothesis is that stripes interfere with optic flow patterns needed by flying insects to execute controlled landings. This could occur through disrupting the radial symmetry of optic flow via the aperture effect (i.e. generation of false motion cues by straight edges), or through spatiotemporal aliasing (i.e. misregistration of repeated features) of evenly spaced stripes. By recording and reconstructing tabanid fly behaviour around horses wearing differently patterned rugs, we could tease out these hypotheses using realistic target stimuli. We found that flies avoided landing on, flew faster near, and did not approach as close to striped and checked rugs compared to grey. Our observations that flies avoided checked patterns in a similar way to stripes refutes the hypothesis that stripes disrupt optic flow via the aperture effect, which critically demands parallel striped patterns. Our data narrow the menu of fly-equid visual interactions that form the basis for the extraordinary coloration of zebras.</p>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.