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.
1,506
datasets available to search
ShareScore release 0.9.0
Dataset results
1,506 results for “objects”
CWL run of Alignment Workflow (CWLProv 0.6.0 Research Object)
<p>This dataset folder is a <strong>CWLProv Research Object</strong> that captures the <a href="http://commonwl.org/">Common Workflow Language</a> execution provenance, see <a href="https://w3id.org/cwl/prov/0.6.0">CWLProv 0.6.0</a> or use the <a href="https://pypi.org/project/cwlprov/">cwlprov Python tool</a> to explore.</p> <p> </p> <p>The CWL <a href="https://w3id.org/cwl/view/git/923ae325fa3a6d3ee3871539c65afa06ac86ac54/aligner/sbg-alignment-cwl/topmed-alignment.cwl">alignment workflow</a> included in this case study is designed by <a href="https://github.com/DataBiosphere">Data Biosphere</a>. It adapts the alignment pipeline originally developed at Abecasis Lab, The University of Michigan. This <a href="https://github.com/FarahZKhan/topmed-workflows/tree/cwlprov_testing">workflow</a> is part of NIH Data Commons initiative and comprises of four stages.</p> <p>First step, <strong><em>Pre-align</em></strong><em>,</em> accepts a <em>Compressed Alignment Map</em> (CRAM) file (a compressed format for BAM files developed <em>by European Bioinformatics Institute</em> (EBI)) and human genome reference sequence as input and using underlying software utilities of <strong>SAMtools</strong> such as <em>view</em>, <em>sort</em> and <em>fixmate</em> returns a list of fastq files which can be used as input for the next step.</p> <p>The next step <strong><em>Align</em></strong> also accepts the human reference genome as input along with the output files from <em>Pre-align</em> and uses <strong>BWA-mem</strong> to generate aligned reads as BAM files. <strong>SAMBLASTER</strong> is used to mark duplicate reads and <strong>SAMtools</strong> view to convert read files from SAM to BAM format.</p> <p>The BAM files generated after <em>lign</em> are sorted with <strong><em>SAMtool sort</em></strong>'.</p> <p>Finally, these sorted alignment files are merged to produce single sorted BAM file using SAMtools merge in <strong><em>Post-align</em></strong> step.</p> <p> </p> <p><strong>Steps to reproduce</strong></p> <p>This analysis was run using a 16-core Linux cloud instance with 64GB RAM and pre-installed docker.</p> <ol> <li> <p>Install <strong>gsutils</strong><br> </p> <pre><code class="language-bash">export CLOUD_SDK_REPO="cloud-sdk-$(lsb_release -c -s)" echo "deb http://packages.cloud.google.com/apt $CLOUD_SDK_REPO main" | \ sudo tee -a /etc/apt/sources.list.d/google-cloud-sdk.list curl https://packages.cloud.google.com/apt/doc/apt-key.gpg | \ sudo apt-key add - sudo apt-get update && sudo apt-get install google-cloud-sdk</code></pre> <p> </p> </li> <li> <p>Get the data and make the analysis environment ready:<br> </p> <pre><code class="language-bash">git clone https://github.com/FarahZKhan/topmed-workflows.git cd topmed-workflows git checkout cwlprov_testing cd aligner/sbg-alignment-cwl # this is a custom script download google bucket files from json files and create a local json # it needs gsutil to be installed though git clone https://github.com/DailyDreaming/fetch_gs_frm_json.git # Wait... this should download ~18Gb. python2.7 fetch_gs_frm_json/dl_gsfiles_frm_json.py topmed-alignment.sample.json </code></pre> <p> </p> </li> <li> <p>Run the following commands to create the CWLProv Research Object:</p> <pre><code class="language-bash">time cwltool --no-match-user --provenance alignmnentwf0.6.0 --tmp-outdir-prefix=/CWLProv_workflow_testing/intermediate_temp/temp --tmpdir-prefix=/CWLProv_workflow_testing/intermediate_temp/temp topmed-alignment.cwl topmed-alignment.sample.json.new zip -r alignment_0.6.0_linux.zip alignment_0.6.0_linux sha256sum alignment_0.6.0_linux.zip > alignment_0.6.0_linux.zip.sha25</code></pre> <p> </p> </li> </ol>
Indoor Object Detection Dataset
<p>We introduce a new fully labeled object detection dataset collected from indoor scenes. This indoor dataset consists of 2213 image frames containing seven classes. In contrast to existing indoor datasets, our dataset includes a variety of background, lighting conditions, occlusion and high inter-class differences.<br> For detail information, please refer to our paper: <a href="https://doi.org/10.1109/EUVIP.2018.8611732">10.1109/EUVIP.2018.8611732</a></p>
Neutrino Quantum Kinetics in Compact Objects
<p>This contains the codebase and data from "Neutrino Quantum Kinetics in<br> Compact Objects" by Sherwood Richers, Gail McLaughlin, Alexey<br> Vlasenko, and James P. Kneller. (2019, submitted to PrD)</p> <p>======================<br> All f.h5 file contents<br> ======================<br> r(cm) - proxy for the time (r=ct)<br> dr_block(cm) - proxy for the block timestep (dr=cdt). Stored for<br> recovery purposes<br> dr_int(cm) - proxy for the interaction integration timestep<br> (dr=cdt). Stored for recovery purposes.<br> dr_osc(cm) - proxy for the oscillation integration timestep<br> (dr=cdt). Stored for recovery purposes.<br> fmatrixf - dimensionless distribution function. Accessed with<br> [time][helicity][energy][flavor1][flavor2][0=real,1=imaginary]</p> <p>==================<br> Directory contents<br> ==================</p> <p>fiducial - simulation output from the fiducial case of a background<br> with rho=1e10g/ccm, T=10MeV, Ye=0.3, the HShen EOS, including<br> oscillations, neutrino absorption on nucleons and nuclei, inelastic<br> electron scattering, elastic nucleon scattering, electron-positron<br> pair production, effective-absorption nucleon-nucleon bremsstrahlung,<br> and neutrino-neutrino scattering and pair processes. Using 25 energy<br> bins covering up to 100 MeV.</p> <p>noosc_full - simulation output from the fiducial case with the same<br> circumstanses as above, except without oscillations and using 50<br> energy bins up to 100 MeV.</p> <p>noosc_supernova_background.dat - snapshot from a 1D core-collapse<br> supernova simulation. The only relevant<br> quantities are the radius (col. 2), density<br> (col. 3), electron fraction (col. 4),<br> temperature (col. 5), electron chemical<br> potential (col. 9), proton chemical potential<br> (col. 10), and neutron chemical potential<br> (col. 11)<br> <br> noosc_*_tdecohere.dat - decoherence time (col. 4) for neutrinos at<br> every radial point (col. 1), each helicity<br> (col. 2), and every energy (col. 3), using 50<br> energy bins at integer multiples of 2 MeV.</p> <p> full - calculations done including the full set of<br> collision interactions listed in fiducial.</p> <p> nonu4 - calculations done including the full set of<br> collision interactions listed in fiducial<br> except neutrino-neutrino scattering/pair<br> processes.</p>
SocRob's ERL specific objects dataset
<p>This dataset contains the images obtained through data augmentation of segmented objects.</p> <p>Used to train the specific object detector for the following classes: coke, water, milkshake, juice, tea, apple, lemon, pear, rice, macaroni, cookies, cloth, toilet_paper, sponge, soap, cup, bowl.</p> <p>In contains augmented images (10000), source images of the objects and their segmentation labels, background images and the projection of the support surfaces where the object patches are placed in the synthesizing process.</p>
Replication Package for MORCoRA: Multi-Objective Refactoring Recommendation Considering Review Availability
<div> <div>This is the replication package for <em>MORCoRA: Multi-Objective Refactoring Recommendation Considering Review Availability</em></div> <br><br> <div> <p><strong>Refactoring sequences searched by MORCoRA</strong></p> </div> <div>The 6 directories contain the refactoring sequences searched by MORCoRA.</div> <div>The name of each directory is the name of the repository introduced in <em>Table 2. Dataset</em>.</div> <br> <div>Each directory includes 6 CSV files.</div> <div>The name of the CSV file represents the search algorithm used to search refactoring sequences.</div> <br> <div>Note that <em>NsgaiiN</em> represents using the NSGA-II algorithm without considering the review availability objective, which is the RA- in <em>Section 4.5</em>.</div> <br> <div>In each CSV file, a single row is the refactoring sequences searched and the recommended reviewers for it. The sequence with no appropriate reviewers will be noted as <em>No appropriate expertise reviewer</em>.</div> <br><br> <div>In each CSV file, a row consists of multiple columns, the last column represents the recommended reviewer, and the rest columns represent elements in the refactoring sequences.</div> <div>Each element consists of:</div> <div> <ul> <li>ROType: refactoring operation type</li> </ul> </div> <div> <ul> <li>class1info: <em>class1</em> in the <em>Table 1. Refactoring Operations</em> in the paper. It is the information of the source class where the number before <em>#</em> represents the access modifiers according to <a href="https://docs.oracle.com/javase/7/docs/api/constant-values.html#java.lang.reflect.Modifier.PROTECTED">modifiers and their corresponding int values</a>, the name after <em>#</em> represents the name for the class</li> </ul> </div> <div> <ul> <li>class1path: the path to the file containing the class</li> </ul> </div> <div> <ul> <li>class2info: <em>class2</em> in the <em>Table 1. Refactoring Operations</em> the formation of the target class</li> </ul> </div> <div> <ul> <li>class2path: the path to the file containing the class</li> </ul> </div> <div> <ul> <li>target: it can be "class" or "method" or "field" according to the refactoring type. The number before <em>#</em> represents the access modifier. The name after <em>#</em> is the name of the "class" or "method" or "field", and its type is revealed after the <em>@</em> symbol.</li> </ul> </div> <br><br> <div><strong>Manually review results</strong></div> <div>The manual review results of the 60 solutions introduced in <em>Section 4.3</em> is recorded in the <em>manually_review_60_solutions.csv</em></div> <br> <div>It includes 6 columns:</div> <div> <ul> <li>Repository: The name of the repository.</li> </ul> </div> <div> <ul> <li>Recommended Refactoring Operations: the searched refactoring sequence.</li> </ul> </div> <div> <ul> <li>Recommended Reviewer: the reviewer recommended to review the refactoring sequence.</li> </ul> </div> <div> <ul> <li>Reviewable: If no appropriate expertise reviewer is found for the sequence, the value is "0", otherwise "1".</li> </ul> </div> <div> <ul> <li>Code smell Eliminated: The code smell type detected by JDeodorant that the recommended refactoring can eliminate. If the refactoring sequence cannot eliminate any code smell, then it is "No".</li> </ul> </div> <div> <ul> <li>Valid: If the refactoring sequence is recommended with appropriate reviewer (value in the column "Reviewable" is "1") and meaningful, and eliminates at least one code smell, the value is "1", otherwise "0".</li> </ul> </div> </div>
User Generated Content (EOL v2): curation of media objects
trust, untrust, hide, set as exemplar <p></p>https://eol-jira.bibalex.org/browse/DATA-1731 Oct 24, 2018 For questions or use cases calling for large, multi-use aggregate data files, please visit the EOL Services forum at <p></p>http://discuss.eol.org/c/eol-services
BioImages - the Virtual Fieldguide (UK): Malcolm Storey text objects
Archival; not in use
Inlists for "Rethinking Thorne-Żytkow Object Formation: The Fate of X-ray Binary LMC X-4 and Implications for Ultra-long Gamma-ray Bursts"
<div> <div> <div> <p>We use the MESA Isochrones and Stellar Tracks (MIST) package (Dotter 2016; Choi et al. 2016) with MESA v7503 (Paxton et al. 2011, 2013, 2015) and mesasdk x86_64-linux-20141212 revision 245. </p> </div> </div> </div>
TUD-GV Dataset for Floating Litter Detection (object detection task)
<p>This dataset contains the data used for the publication:</p> <p>Jia, T., de Vries, R., Kapelan, Z., van Emmerik, T. H., & Taormina, R. (2024). Detecting floating litter in freshwater bodies with semi-supervised deep learning. <em>Water Research</em>, <em>266</em>, 122405.</p> <p>This dataset is a subset of the large-scale "TU Delft - Green Village" (TUD-GV), which includes 9,473 RGB images. More details on the TUD-GV dataset can be found at: <a href="https://doi.org/10.5281/zenodo.7636124" target="_blank" rel="noopener">https://doi.org/10.5281/zenodo.7636124</a>. This subset used in this publication consists of 1,501 images, selected from the full TUD-GV dataset. All floating litter items in this subset have been annotated with bounding boxes. This subset is specifically for detecting floating litter in object detection tasks.</p> <p>The 1,501 images are stored in the <em>images.zip</em> file, the annotations are stored in the <em>labels_txt.zip</em> file, and the class of the annotation (i.e., litter) is stored in the <em>classes.txt</em> file.</p> <p>If you use this dataset for a publication, please cite the paper. Here is a BibTeX entry:</p> <pre>@article{jia2024detecting, title={Detecting floating litter in freshwater bodies with semi-supervised deep learning}, author={Jia, Tianlong and de Vries, Rinze and Kapelan, Zoran and van Emmerik, Tim HM and Taormina, Riccardo}, journal={Water Research}, volume={266}, pages={122405}, year={2024}, publisher={Elsevier} }</pre>
The double-peaked Type Ic supernova 2019cad: another SN 2005bf-like object
<p>SN 2019cad spectra and bolometric LC.</p> <p>Paper available at: https://academic.oup.com/mnras/article/504/4/4907/6224870</p>
Yolo object detector raw results for example manipulations
<p>There is one example of the direct object detection outputs from YOLO for each manipulation type. Should you sish for more datasets, please contact the authors</p>
Simulated Object-Centric Event Logs (OCEL 2.0) for Order-to-Cash, Procure-to-Pay, Hiring, and Hospital Patient Lifecycle Processes
<p>This dataset contains simulated object-centric event logs for four distinct business processes: <strong>Order-to-Cash (O2C)</strong>, <strong>Procure-to-Pay (P2P)</strong>, <strong>Hiring</strong>, and <strong>Hospital Patient Lifecycle</strong>. Each process is designed to reflect realistic workflows, encompassing multiple object types and capturing key activities, decision points, and process dynamics. The dataset is aimed at providing a rich source of data for process mining, analysis, and modeling activities.</p> <p>1. <strong>Order-to-Cash (O2C)</strong>:<br> The O2C process simulates an end-to-end business flow starting from customer order placement to payment receipt. It includes diverse activities such as order approval, fulfillment, invoice generation, and payment processing, involving object types like Customers, Orders, Products, and Invoices. The dataset captures variability through random decisions, synchronization between departments, and workarounds in credit checks and inventory adjustments. Attributes such as customer tiers, order values, and shipment statuses add further depth, allowing for detailed analysis of this complex process.</p> <p>2. <strong>Procure-to-Pay (P2P)</strong>:<br> The P2P process simulates the procurement lifecycle, from requisition creation to payment of suppliers. Key activities include purchase order creation, three-way matching, goods receipt, and payment processing. The event log records object types such as Purchase Requisitions, Purchase Orders, Suppliers, and Invoices. Variability is introduced through approval decisions, batching, and potential mismatches in the matching process. The dataset represents the inherent complexities of real-world procurement operations, including batching and synchronization issues between different process stages.</p> <p>3. <strong>Hiring Process</strong>:<br> The hiring process log tracks the recruitment lifecycle, from job requisition creation to onboarding. It includes object types like Candidates, Job Requisitions, Recruiters, and Interviewers. The process covers activities such as resume screening, interviews, assessments, and offer management. Variability in the hiring process is introduced through random delays, candidate decisions, and background check durations. Batching occurs in stages like resume screening and onboarding, while synchronization challenges arise during interview scheduling.</p> <p>4. <strong>Hospital Patient Lifecycle</strong>:<br> This log represents the lifecycle of patients within a hospital, capturing interactions with multiple resources such as physicians, beds, and medical equipment. The process begins with pre-admission activities, followed by diagnosis, treatment, and discharge. The dataset includes object types like Patients, Physicians, and Medical Equipment, with attributes related to patient demographics and event severity. The process reflects the dynamic nature of hospital operations, including synchronization of resources and the occurrence of workarounds in case of delays or resource unavailability.</p> <p>Each process simulation captures high variability, synchronization issues, and batching, making this dataset suitable for analyzing real-world operational challenges. The logs provide a comprehensive view of complex workflows, supporting advanced analysis, including object-centric process mining.</p> <p>This description will provide the necessary details about the dataset, highlighting its structure, purpose, and potential uses for researchers and process analysts.</p> <p>Object-centric event logs conceived and simulated by the <strong>o1-preview-2024-09-12</strong> LRM, using the https://github.com/fit-alessandro-berti/llm-ocel-simulator project.</p> <p> </p> <p> </p>
Path Relinking Strategies for the bi-objective Double Floor Corridor Allocation Problem
<p>Dataset, code and results for the "Path Relinking Strategies for the bi-objective Double Floor Corridor Allocation Problem" paper</p>
Dump truck object detection with manual annotations
<p>Doing manual annotations can sometimes be resource heavy, depending on the amount of data. This dataset was designed to created to use in conjunction with a semi-automatic annotation method based on linear interpolation. The dataset contains 799 images, where 679 lies in the trainingset, and the rest lies in the validationset. The images are taken from 6 different video streams, where a remote controlled wheel loader approaches a miniature dump truck at different angles. 4 of the videos are used in the trainingset. The labels can contain up to 5 classes which are:</p> <p>0 - front wheel <br> 1 - middle wheel<br> 2 - back wheel<br> 3 - tipping body<br> 4 - cap</p> <p>This dataset was used to train a YOLOv3 model, hence the labels will be written in the YOLO labeling format.</p>
Event-driven Proto-object based saliency in 3D space to attract a robot's attention - Supplementary Video
<p>To interact with its environment, a robot working in 3D space needs to organise its visual input in terms of objects or their<br> perceptual precursors, proto-objects. Among other visual cues, depth is a submodality used to direct attention to visual features<br> and objects. Current depth-based proto-object attention models have been implemented for standard RGB-D cameras that<br> produce synchronous frames. In contrast, event cameras are neuromorphic sensors that loosely mimic the function of the<br> human retina by asynchronously encoding per-pixel brightness changes at very high temporal resolution, thereby providing<br> advantages like high dynamic range, efficiency (thanks to their high degree of signal compression), and low latency. We<br> propose a bio-inspired bottom-up attention model that exploits event-driven sensing to generate depth-based saliency maps<br> that allow a robot to interact with complex visual input. We use event-cameras mounted in the eyes of the iCub humanoid robot<br> to directly extract edge, disparity and motion information. Real-world experiments demonstrate that our system robustly selects<br> salient objects near the robot in the presence of clutter and dynamic scene changes, for the benefit of downstream applications<br> like object segmentation, tracking and robot interaction with external objects.</p>
Object Detection meets Knowledge Graphs Raw Data
<p>Raw data used to reproduce the results of 'Object Detection meets Knowledge Graphs'. Combination of VOC 2007 images, MS COCO 2014 images and ConceptNet knowledge graph. Also the checkpoints of the two trained models for the VOC dataset and COCO dataset.</p>
Ground truth 3d F-object
<p>This dataset contains two 3D model files of a support test object in the shape of a 3D F, with a sharp point. The 3D model files differ in their scaling.</p> <p>The files in this dataset were altered from the Support Test Object by feilen available from <a href="https://www.thingiverse.com/thing:4360">https://www.thingiverse.com/thing:4360</a> .</p>
Instance data: open-shop scheduling problems with any regular minsum objective
<p>Open shop scheduling instances used in the working paper</p> <ul> <li>Emde, S. & Lysgaard, J. (2021). Branch-cut-and-price for open-shop scheduling problems with any regular minsum objective.</li> </ul> <p>The instances in file <em>Brucker_instances.csv</em> are based on the test data for the classic [O||Cmax] problem from Brucker, P., Hurink, J., Jurisch, B., & Wöstmann, B. (1997). A branch & bound algorithm for the open-shop problem. <em>Discrete Applied Mathematics</em>, <em>76</em>(1-3), 43-59. They are enriched with release and due dates as well as machine-pair dependent transportation delays. The instances in file <em>random_instances.csv </em>are new.</p> <p>The files contain comma-separated values. Each line (except for the header) stands for one instance. The columns labels are:</p> <p>ID: label (identifier) of the instance</p> <p>n: number of jobs</p> <p>m: number of machines</p> <p>p: processing times; each square bracket stands for one machine, the values inside the brackets for the processing times of the jobs on the respective machine</p> <p>t: transfer times between machines; note that the dummy machine 0 where all jobs originate and end is the last machine, i.e., it has the highest index</p> <p>r: release dates for each job</p> <p>d: due dates for each job.</p> <p>Mj: set of machines on which the jobs must be processed; each square bracket stands for one job, values inside the brackets for machine indices. Note that indices are zero-based.</p> <p>In instance set <em>Brucker_instances.csv</em>, every machine is visited by every job, therefore there is no column Mj. Conversely, due dates are immaterial for the random instances because of their quadratic completion time objective. Hence, they are omitted from the table.</p>
Building Object and Outdoor Scene Segmentation (BOOSS) - Multi-channel (RGB + Thermal) Aerial Imagery Datasets
<p>The dataset of <em>Building Object and Outdoor Scene Segmentation (BOOSS)</em> is based on multi-channel aerial imagery data. It covers </p> <p>- Ground Truth</p> <p>- RGB</p> <p>- Thermal</p> <p>The annotations in version 1.0 include roofs, facades, cars, roof equipment, and ground equipment</p> <p>Please cite as:</p> <p>Hou, Yu, Meida Chen, Rebekka Volk, and Lucio Soibelman. "An Approach to Semantically Segmenting Building Components and Outdoor Scenes Based on Multichannel Aerial Imagery Datasets." <em>Remote Sensing</em> 13, no. 21 (2021): 4357.</p>
Deep learning object detection to estimate the nectar sugar mass of flowering vegetation
<p>Floral resources are a key driver of pollinator abundance and diversity, yet their quantification in the field and laboratory is laborious and requires specialist skills.</p> <p>Using a dataset of 25000 labelled tags of fieldwork-realistic quality, a Convolutional Neural Network (Faster R-CNN) was trained to detect the nectar-producing floral units of 25 taxa in surveyors' quadrat images of native, weed-rich grassland in the UK.</p> <p>Floral unit detection on a test set of 50 model-unseen images of comparable vegetation returned a precision of 90%, recall of 86% and F1 score (the harmonic mean of precision and recall) of 88%. Model performance was consistent across the range of floral abundance in this habitat. </p> <p>Comparison of the nectar sugar mass estimates made by the CNN and three human surveyors returned similar means and standard deviations. Over half of the nectar sugar mass estimates made by the model fell within the absolute range of those of the human surveyors.</p> <p>The optimal number of quadrat image samples was determined to be the same for the CNN as for the average human surveyor. For a standard quadrat sampling protocol of 10–15 replicates, this application of deep learning could cut pollinator-plant survey time per stand of vegetation from hours to minutes.</p> <p>The CNN is restricted to a single view of a quadrat, with no scope for manual examination or specimen collection, though in contrast to human surveyors its object detection is deterministic and floral unit definition is standardised.</p> <p>As agri-environment schemes move from prescriptive to results-based, this approach provides an independent barometer for grassland management which is usable by both landowner and scheme administrator. The model can be adapted to visual estimations of other ecological resources such as winter bird food, floral pollen volume, insect infestation and tree flowering/fruiting, and by adjustment of classification threshold may show acceptable taxonomic differentiation for presence-absence surveys.</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.