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.
250
datasets available to search
ShareScore release 0.9.0
Dataset results
250 results for “Synthetic dataset”
Real and Synthetic Dataset for Active Shooter Situations
<p>This dataset includes annotations for shooters (both person and visible portion of the gun) and guns in a primarily indoor setting. The synthetic data was generated using Unreal Engine 4 and Unreal Engine 5 and contains both the default semi-realistic textures of the environments and segmentation masks as a form of domain randomization. The textured synthetic data was further augmented with camera sensor effects as another domain adaptation technique. The folders use the following notation: T denotes textured synthetic, M denotes masked synthetic, R denotes real, and the prefix A denotes augmented. We also include some annotated videos for evaluating tracking performance.</p>
I-BiDaaS - CRF - Aluminium die-casting Synthetic Dataset
<p>The dataset has been generated after receiving unstructured sets of a large amount of heterogeneous data from several sources and levels from the production line. CRF analysed all information and selected seventeen parameters (e.g. piston speed in the first and second phase, piston stroke, intensification pressures) of the production of the engine block by die-casting. The synthetic dataset was generated in order to analyse the provided parameters and try to cluster them by identifying those that are most representative of each cluster. The clusterization is useful to identify the behaviour of parameters and help to understand those that affect the quality of the process and products. </p>
Synthetic Patient Appointment Dataset
<p>A synthetic dataset of patient appointments, referrals, and journeys to a fictional service in the North East of England. The code can be adjusted to incorporate any area on mainland Great Britain. NI or the islands can be integrated too, however the structure of postcode, GP and OSA public data is different, and data input handlers will need to be adjusted.</p> <p>The behaviour of the patients (visiting their nearby GP followed by attending a</p> <p>specialist clinic), appointments (clinic appointments within 7day-6weeks of the referral (gp appointment)), and facilities (one major facility taking the load, along with minor facilities) is meant to mirror the real data used under Pilot 2 of the Track & Know Project.</p> <p>Real postcodes, from Royal Mail, are used to generate the appointment population, real facilities are used based on the British Lung Foundations study of Obstructive Sleep Apnoea, and real GP's are used based on public data from the NHS.</p>
Large Spots DeepMIB project, synthetic dataset for testing 2D semantic segmentation
<p>A complete DeepMIB project with a synthetic dataset generated for quick tests of semantic segmentation approaches.<br>The dataset includes a trained DeepLabV3-Resnet18 network for detection of large spots on a black background. </p><p>The network can be opened by loading "2D_LargeSpots_2cl_DeepLabV3.mibCfg" file by</p><ul><li><i>MIB->Menu->Tools->Deep learning segmentation->Options tab->Config files->Load </i></li><li>Drag and drop of the config file into DeepMIB window</li></ul><p>Microscopy Image Browser: <a href="https://mib.helsinki.fi">https://mib.helsinki.fi</a></p>
Synthetic dataset for end-to-end Relation Extraction of relationships between Organisms and Natural-Products with Vicuna-13b-v1.5
<p>A new synthetic dataset (training/validation) for end-to-end Relation Extraction of relationships between Organisms and Natural-Products.</p><p>The new dataset was generated using <a href="https://huggingface.co/lmsys/vicuna-13b-v1.5">Vicuna-13b-v1.5</a>, derived from LLaMA 2. Like the model, the produced synthetic data are also submitted to the License of the model used for generation, see the original <a href="https://github.com/facebookresearch/llama/blob/main/LICENSE">LLaMA 2 license</a>.</p><p>The new dataset was created based on the top-1000 (per biological kingdom) LOTUS literature references extracted with the <a href="https://github.com/idiap/gme-sampler">GME-sampler</a>.</p><p>The dataset contains 10,405 items in the training set and 547 items in the validation set.</p><p>The dataset was generated using the same protocol as described in the <a href="https://github.com/idiap/gme-sampler">article</a>.</p>
North Atlantic synthetic tropical cyclone track, intensity, and rainfall dataset from RAFT
<p>The Risk Analysis Framework for Tropical Cyclones (RAFT)'s comprehensive and unified simulation of 40,000 synthetic North Atlantic tropical cyclone (TC) events are presented in this dataset. RAFT meticulously models these events based on large-scale environmental conditions, providing a valuable tool for in-depth TC impact analysis. The dataset encompasses detailed 6-hourly track information, along-track intensity metrics (including maximum wind speed and minimum pressure), the radius of maximum winds, and cumulative precipitation for each event.</p> <p>The primary dataset is encapsulated in a NetCDF4 file, "<a href="../records/10392725/files/RAFT.NA.v20231016.nc?download=1">RAFT.NA.v20231016.nc</a>", which contains a complete array of variables pertinent to the 40,000 synthetic TCs. These variables, detailed in Table 1 of the accompanying paper and summarized below, offer a comprehensive view of each TC event:</p> <ul> <li><strong>Basin ID</strong>: Identifies the basin (1 for North Atlantic)</li> <li><strong>Storm ID</strong>: Unique identification number for each TC, starting from 0</li> <li><strong>Year</strong>: Year of the environmental conditions used for modeling</li> <li><strong>Jday</strong>: Julian day of the year, ranging from 0 to 365</li> <li><strong>Longitude (lon)</strong>: Geographical longitude in degrees</li> <li><strong>Latitude (lat)</strong>: Geographical latitude in degrees</li> <li><strong>Maximum Wind Speed (vmax)</strong>: Measured in knots</li> <li><strong>Minimum Pressure (mslp)</strong>: Measured in hectopascals (hPa)</li> <li><strong>Radius of Maximum Wind (rmax)</strong>: Measured in nautical miles (nmi)</li> </ul> <p>Additionally, the dataset offers individualized accumulated rainfall data for each TC event, stored in NetCDF4 files named according to the convention "modeled_rainfall_ERA5_syn_{i}.h5", where "{i}" is the synthetic storm's ID. "ERA5" signifies the reanalysis input source, and "syn" indicates a synthetic track. This component of the dataset includes the following variables, all measured in total millimeters of precipitation:</p> <ul> <li><strong>Total Accumulated Rainfall (p_accum)</strong></li> <li><strong>Frictional Precipitation Component (p_accum_f)</strong></li> <li><strong>Topographic Precipitation Component (p_accum_h)</strong></li> <li><strong>Shear-related Precipitation Component (p_accum_s)</strong></li> <li><strong>Vortex Stretching Precipitation Component (p_accum_t)</strong></li> </ul> <p>The rainfall dataset is curated to focus on TC events within 600 km of the U.S. coast, reducing the number of rainfall events to 17,010 from the original 40,000, thereby enhancing its relevance and manageability. For user convenience, these events are compressed into grouped archives named "RAFT_accum_rainfall_{index}.tar.gz", where each "{index}" represents the index of the zipfile, containing up to 2,000 files for efficient data retrieval.</p> <p>The accumulated rainfall data is provided on a regular spatial grid, detailed in "<a href="../records/10392725/files/RAFT_rainfall_latlon_grid.h5?download=1">RAFT_rainfall_latlon_grid.h5</a>", which outlines the grid coordinates ('lat' and 'lon').</p> <p>For comprehensive usage guidelines and further insights into this dataset, users are encouraged to refer to the associated paper. This dataset is not only a significant resource for researchers and analysts in the field of meteorology but also serves as a pivotal tool for understanding and predicting the impacts of tropical cyclones.</p> <p> </p> <p><strong>How to cite:</strong></p> <p>Xu, W., Balaguru, K., Judi, D.R. <em>et al.</em> A North Atlantic synthetic tropical cyclone track, intensity, and rainfall dataset. <em>Sci Data</em> <strong>11</strong>, 130 (2024). https://doi.org/10.1038/s41597-024-02952-7</p>
Task-driven neural network models predict neural dynamics of proprioception: Synthetic muscle spindle datasets
<p>#############</p> <p>Task-driven neural network models predict neural dynamics of proprioception, Cell 2024</p> <p>#############</p> <p>Authors: Marin Vargas, Alessandro (orcid=0000-0001-7073-4120) and Bisi, Axel (orcid=0009-0006-8602-7555) and Chiappa, Alberto Silvio (orcid=0009-0001-2764-6552) and Versteeg, Christopher (orcid=0000-0002-4269-5109) and Miller, Lee E. (orcid=0000-0001-8675-7140) and Mathis, Alexander (orcid=0000-0002-3777-2202)</p> <p>Affiliation: EPFL</p> <p>Date: January, 2024</p> <p>Link to the Cell article: </p> <p><a href="https://www.cell.com/cell/pdf/S0092-8674(24)00239-3.pdf">https://www.cell.com/cell/pdf/S0092-8674(24)00239-3.pdf</a></p> <p>--------------------------------</p> <p>Here we provide the synthetic spindle datasets of our article "Task-driven neural network models predict neural dynamics of proprioception". It contains the synthetic generated training dataset of simulated muscle spindles during arm passive movements generated with either character writing (PCR) or with 3D target reaching using reinforcement learning (RL).</p> <p>The overall structure of the data is:</p> <p>└── spindle_datasets<br> ├── pcr_dataset - Contains PCR synthetic training dataset<br> └── rl_dataset - Contains RL-generated synthetic training dataset</p> <p>The code to generate the PCR synthetic spindle dataset is available at: <a href="https://github.com/amathislab/Task-driven-Proprioception/tree/master/PCR-data-generation">https://github.com/amathislab/Task-driven-Proprioception/tree/master/PCR-data-generation</a></p> <p>The code to generate the RL-generated synthetic spindle dataset is available at: <a href="https://github.com/amathislab/Task-driven-Proprioception/tree/master/RL-data-generation">https://github.com/amathislab/Task-driven-Proprioception/tree/master/RL-data-generation</a></p> <p>--------------------------------</p> <p>The datasets, weights, activations and predictions are released with Creative Commons Attribution 4.0 license.</p> <p>The code is released under the MIT license, see <a href="https://github.com/amathislab/Task-driven-Proprioception">https://github.com/amathislab/Task-driven-Proprioception</a></p> <p>If you find our code, weights, predictions or ideas useful, please cite:</p> <p>@article{vargas2024task,<br> title={Task-driven neural network models predict neural dynamics of proprioception},<br> author={{Marin Vargas}, Alessandro and Bisi, Axel and Chiappa, Alberto S and Versteeg, Chris and Miller, Lee E and Mathis, Alexander},<br> journal={Cell},<br> year={2024},<br> publisher={Elsevier}<br>}</p>
SDADDS-Guelma : A Multi-purpose Dataset for Synthetic Degraded Arabic Documents
<h1><strong>SDADDS-Guelma : A Multi-purpose Dataset for Synthetic Degraded Arabic Documents </strong></h1> <h2><strong>Description:</strong></h2> <p>This is a partial release of the SDADDS-Guelma dataset.</p> <p>SDADDS-Guelma (Synthetic Degraded Arabic Document DataSet of the University of Guelma) is a database of synthetic noisy or degraded Arabic document images. It was created by Dr. Abderrahmane Kefali and his team to support research on preprocessing, analysis, and recognition of degraded Arabic documents, where having a large set of images for training and testing is essential. This dataset is made publicly available to researchers in the field of document analysis and recognition, with the hope that it will be useful and contribute to their research endeavors.</p> <p>In this first release of the dataset, 84 handwritten images and 120 printed images have been used, along with 25 images of historical backgrounds, forming a total of 26316 synthetic images of degraded Arabic documents along with their corresponding ground-truth files.</p> <p>This release is separated into two parts to facilitate upload and use: one for the handwritten documents and the second for the printed documents.</p> <h2><strong>Composition of the dataset:</strong></h2> <p>Each of the parts of the SDADDS-Guelma dataset is organized into directories as follows:</p> <ul> <li>TXT_Files: Contains texts in UTF-8 format. </li> <li>IMG: Contains images of printed and handwritten Arabic text constructed from the text files. </li> <li>Bin_IMG: Contains binary images corresponding to the original images. </li> <li>BG_IMG: Contains images of empty old document backgrounds used for the generation of synthetic historical document images. </li> <li>GT_Files: Contains XML annotation files corresponding to the text images.</li> <li>Degraded_IMG: This directory contains synthetically generated degraded images, separated into sub-directories based on noise types such as Local_Noise, Show_through, Rotation, Curvature, Comb_IMG, etc.</li> </ul> <h2><strong>Ground-truth information:</strong></h2> <p>Ground truth information is essential for a document dataset, as it annotates documents and represents their essential characteristics. Our dataset is designed to be a large-scale and multipurpose dataset. As such, our methodology ensures that ground truth information is provided at three levels: text level (character codes), pixel level (binary and cleaned image), and document physical structure and other annotation information level.</p> <ul> <li>Textual Ground Truth: these are identical to the original texts. </li> <li>Pixel-level ground truth: presented in the form of binary images.</li> <li>Ground truth at the document structure level: the structure of each document image, alongside the textual transcription of the words and PAWs, is recorded in a corresponding XML annotation file. The XML format utilized resembles that employed in similar works with adjustments made according to the specific characteristics of Arabic texts, including the presence of PAWs. </li> </ul> <p>Consequently, each original text image in our dataset is associated to an XML file detailing the entire ground truth and associated metadata. </p> <h3><em><strong>Structure of XML file:</strong></em></h3> <p>Each XML annotation file contains metadata about the document image and text content within the image, including the language, number of lines, and font attributes. It also provides detailed information about each text line, word, and Part of Arabic Words (PAWs), including their bounding boxes and textual transcriptions.</p> <p>Thus, each ground truth file takes the following form:</p> <pre><code><DOCUMENT imageName="PR1Kufi_bin.png" height="2631" width="1860" nbTextLines="8" language="Arabic" fontName="Kufi" fontSize="34"> <TEXTLINE id="0" nbWords="3" boundingBox="215,481,355,1379"> <WORD id="0" nbPAWs="2" boundingBox="217,1065,341,1379" transcription="خصائص"> <PAW id="0" nbCCs="2" boundingBox="217,1206,322,1379" transcription="خصا"> <CC id="0" nbPixels="4110" pixels="(217,1206,1206);(218,1206,1207);(219,1206,1208);(220,1206,1208);(221,1206,1211);(222,1206,1211);..."> </CC> <CC id="1" nbPixels="80" pixels="(263,1330,1336);(264,1329,1336);(265,1328,1337);(266,1328,1337);(267,1328,1337);(268,1328,1337);(269,1328,1337);...."> </CC> </PAW> .... </WORD> <WORD id="1" nbPAWs="2" boundingBox="215,817,338,1044" transcription="التفسير"> <PAW id="0" nbCCs="1" boundingBox="215,1030,322,1044" transcription="ا"> <CC id="0" nbPixels="1037" pixels="(215,1030,1030);(216,1030,1030);(217,1030,1031);..."></CC> .... </PAW> .... </WORD> </TEXTLINE> .... </DOCUMENT></code></pre> <h1><strong>Contact:</strong></h1> <p>Name: Dr. Abderrahmane Kefali<br>Affiliation: University of 8 May 1945-Guelma, Algeria<br>Email: kefali.abderrahmane@univ-guelma.dz</p>
Synthetic Product Desirability Datasets for Sentiment Analysis Testing
<p><strong>Overview:</strong><br>This collection contains three synthetic datasets produced by gpt-4o-mini for sentiment analysis and PDT (Product Desirability Toolkit) testing. Each dataset contains 1000 hypothetical software product reviews with the aim to produce a diversity of sentiment and text. The datasets were created as part of the research described in:</p> <p>Hastings, J.D., Weitl-Harms, S., Doty, J., Myers, Z. L., and Thompson, W., “Utilizing Large Language Models to Synthesize Product Desirability Datasets,” in Proceedings of the 2024 IEEE International Conference<br>on Big Data (BigData-24), Workshop on Large Language and Foundation Models (WLLFM-24), Dec. 2024.<br><a href="https://arxiv.org/abs/2411.13485">https://arxiv.org/abs/2411.13485</a>.</p> <p>Briefly, each row in the datasets was produced as follows:<br>1) Word+Review: The LLM selected a word and synthesized a review that would align with a random target sentiment.<br>2) Review+Word: The LLM produced a review to align with the target sentiment score, and then selected a word appropriate for the review.<br>3) Supply-Word: A word was supplied to the LLM which was then scored, and a review was produced to align with that score.</p> <p>For sentiment analysis and PDT testing, the two columns of main interest across the datasets are likely 'Selected Word' and 'Hypothetical Review'.</p> <p><strong>License:</strong><br>This data is licensed under the CC Attribution 4.0 international license, and may be taken and used freely with credit given. Cite as:</p> <p>Hastings, J., Weitl-Harms, S., Doty, J., Myers, Z., & Thompson, W. (2024). Synthetic Product Desirability Datasets for Sentiment Analysis Testing (1.0.0). Zenodo. <a href="https://doi.org/10.5281/zenodo.14188456" target="_blank" rel="noopener">https://doi.org/10.5281/zenodo.14188456</a></p>
LLM Generated Synthetic Dataset of DoS Exposed Solidity Contracts
<p>This dataset provides the replication package for the paper 'Large Language Models for Synthetic Dataset<br>Generation: A Case Study on Ethereum Smart Contract DoS Vulnerabilities' accepted for publication at the 8th International Workshop on Blockchain Oriented Software Engineering. The provided sources encompass:<br>1) The synthetic contracts (Vulnerable, Exploit, and Patched contract for each use case) generated by Claude and GPT4.<br>2) The configuration files of the hardhat-based testing environment.<br>3) The test suite that showcases the vulnerabilities of the generated contracts (including mock contracts) (hardhat is required to run and test contracts).<br><br></p> <div> <p> </p> </div>
Synthetic dataset from - Bar et al., Sifting through the haystack - efficiently finding rare behaviors in large-scale datasets, WACV 2025
<p>This is a synthetic dataset emulating pose estimation data, introduced in the associated paper. </p> <p>Briefly, each sample in the dataset is a sequence of 5-keypoints with 9 timesteps. Movement of each keypoint in time is determined by a sinus with some amplitude A and some frequency f, this is loosely inspired by larval zebrafish swimming movement. <br>There are two types of behaviors - a common behavior (aka forming the majority of the samples in the dataset) where the frequency of the sinus is larger than the amplitude, and a rare one where the amplitude is larger than the frequency. <br>We vary the similarity between the rare and common behaviors by relaxing the standard deviation of the gaussian from which we draw these movement parameters (behavior similarity, sd= [0.5, 1.5, 2.5, 5]). We also test different levels of data imbalance, varying the frequency of the rare behavior (rarity=[1.5%,5%,12%,24%]). </p> <p>Thus we created 16 datasets with all possible combinations.</p> <p>The data generation code will become available in our code repository: https://github.com/shir3bar/SiftingTheHaystack</p> <p>The data was used to create a controlled experimental sandbox in which we could test our pipeline for detecting rare behaviors. Sounds interesting? Read our paper and check out the code :)</p> <p> </p> <p> </p>
dataset for "Non-LTE Synthetic Observables of a Multidimensional Model of Type Ia Supernovae"
<p>Included here are datafiles supplementary to the article "Non-LTE Synthetic Observables of a<br>Multidimensional Model of Type Ia Supernovae" (Boos, Dessart, Townsley, Shen), submitted <br>to ApJ/arxiv in October 2024. This is a revised version of the dataset using an improved<br>method to generate the 1D LTE observables (1D LTE and pseudo-2D non-LTE observables have<br>thus changed from the previous version).</p> <p>We include in this dataset the 2D double detonation ejecta model (produced in Boos et al. 2021),<br>as well as the 15 wedge profiles constructed from this model, which were used in the radiative<br>transfer calculations of this work. These profiles are provided in Sedona input format. The<br>headers/columns for the 1D profiles are as follows:<br>1: Sedona model type<br>2: num_cells, rmin, ejecta time, num_isos<br>3: isos<br>4-: v_r, rho, T, X_a, X_b, etc.</p> <p>We also include the synthetic spectra and photometry from this work. These synthetic observables<br>are given for each set of calculations (1D LTE, 2D LTE, 1D non-LTE, and pseudo-2D non-LTE) between<br> -3 and +15 d from maximum light. The photometry is given in absolute magnitude in the Vega system.<br>The spectra have been rebinned such that the bins are the same between the Sedona and CMFGEN<br>calculations. The flux in these spectral files is given for a distance of 1 kpc.</p> <p><br>-Samuel J. Boos (sjboos@crimson.ua.edu)</p>
Synthetic dataset of a full wavefield representing the propagation of Lamb waves and their interactions with delaminations
<p>The dataset contains 475 simulated cases of full wavefield of Lamb waves propagation in a plate made of carbon fibre reinforced plastic (CFRP). The simulated 475 cases represent delaminations with different locations, shapes, and sizes. The following random factors were simulated in each case:</p> <ul> <li>delamination geometrical size (ellipse minor and major axis randomly selected from the interval [10 mm, 40 mm],</li> <li>delamination angle (randomly selected from the interval [0◦ , 180◦]),</li> <li>coordinates of the centre of delamination (randomly selected from the interval [0 mm, 250 mm − δ] and [250 mm + δ, 500 mm], where δ = 10 mm).</li> </ul> <p>The guided waves were excited at the centre of the plate by applying equivalent piezoelectric forces. The excitation was in the form of toneburst sine signal modulated by the Hann window. The carrier frequency is assumed 50 kHz, and the modulation frequency is 10 kHz. The total wave propagation time was set to 0.75 ms so that the guided wave can propagate to plate edges and back to the actuator twice. The number of time integration steps was 150000 which was selected for the stability of the central difference scheme.</p> <p>The material is a typical cross-ply CFRP laminate. The stacking sequence [0/90]<sub>4</sub> was used in the model. The properties of a single-ply were as follows [GPa]: C<sub>11</sub> = 52.55, C<sub>12</sub> = 6.51, C<sub>22</sub> = 51.83, C<sub>44</sub> = 2.93, C<sub>55</sub> = 2.92, C<sub>66</sub> = 3.81. The assumed mass density was 1522.4 kg/m3. These properties were selected so that simulated numerically wave front patterns and wavelengths are similar to the wavefields measured by SLDV on CFRP specimens used later on for testing the developed methods for delamination identification. The shortest wavelength of propagating A0 Lamb wave mode was 21.2 mm for numerical simulations and 19.5 mm for experimental measurements.</p> <p><br> In each delamination case, 512 frames were generated to visualise the propagation of Lamb waves and their interactions and reflections from the delamination and edges.<br> The numerically generated dataset resembles the velocity measurements acquired by the scanning laser Doppler vibrometer (SLDV) at the bottom surface of the plate of dimensions 500×500 mm.</p> <p>The uploaded dataset contains two ZIP files:</p> <ol> <li> The first file contains 475 folders regarding all cases of different delaminations and their interaction with Lamb waves. In each case, there are 512 images in PNG format representing the propagation of guided waves.</li> <li> The second file contains: <ul> <li>475 images in PNG format representing the ground truth of the delaminations.</li> <li>CSV file contains all info regarding delaminations.</li> </ul> </li> </ol> <p> </p>
Synthetic dataset for the flow-shop problem with sequence-dependent setup times (FSsdsu_DS_01)
<p>Synthetic dataset containing setup and processing times for a flow-shop problem of 1.000 parts and 200 machines. The setup times are specified for each machine taking into account the previous part produced on the machine.</p>
Synthetic dataset for the flow-shop problem (FS_DS_01)
<p>Synthetic dataset containing setup and processing times for a flow-shop problem of 1.000 parts and 200 machines.</p>
deepNIR: Dataset for generating synthetic NIR images and improved fruit detection system using deep learning techniques
<p>In this paper, we present datasets that can be utilised for synthetic near infrared (NIR) image and bounding box level fruit detection system. It is undeniable fact that high-caliber machine learning software frameworks such as Tensorflow or Pytorch and large scale dataset such as ImageNet and COCO, and accelerated GPU hardware support have pushed the limit of machine learning for more than decades.</p> <p>Among these breakthroughs quality dataset is one of important key building blocks that can lead to success in model generalisation and deployment for data-driven deep neural networks. Particularly, synthetic data generation such as generative adversarial networks often requires relatively larger scale data than other supervised approaches. In addition, posing constrains such as geometrical facial constrains in fake face generation or consistent and radiometrically calibrated reflectances from satellite imagery commonly yield better results. We share NIR+RGB dataset that are re-processed from other two public datasets (nirscene and SEN12MS) and our own novel sweetpepper dataset to be able to timely adopt to other following studies.</p> <p>We oversampled from original nirscene dataset at 10, 100, 200, and 400 ratios and total of 127k pair of images. For SEN12MS satellite multispectral dataset, we selected one largest subset; Summer (45k) and All seasons (180k). Our sweetpeppr dataset consists of 1,615 pairs of NIR+RGB images. We demonstrate these NIR+RGB datasets are sufficient to be used for synthetic NIR generation quantitatively and qualitatively. We achieved Frechet Inception Distance (FID) of 11.36, 26.53, and 40.15 for nirscene1, SEN12MS, and sweetpepper dataset respectively.</p> <p>We also release <em>11</em> fruits' bounding box annotations that can be exported as various formats using cloud service. 4 newly added fruits [blueberry, cherry, kiwi, and wheat] compounds 11 novel bounding box dastaset together with our previous work in deepFruits project [apple, avocado, capsicum, mango, orange, rockmelon, strawberry]. The total number of bounding box instances is 162k and all bounding box dataset is ready for use from cloud service. For evaluation of these dataset, Yolov5 single stage detector is exploited and reported impressive mean-average-precision, mAP[0.5:0.95] results of [min:0.49, max:0.812]. We hope these dataset is useful and serves as one of baseline for the following up studies.</p>
6Dof_synthetic_dataset
<p>The dataset consists of a set of computer-generated images for the four objects (TV frame, green PCB, yellow PCB, and screwdriver) involved in the TV assembly task of CoLLaboratE. For every frame a pair of rgb and depth image, the mask for each object, object poses and the corresponding camera pose are stored. In total there are 3700 different poses for every object, and in addition to that 17 different backgrounds were used to augment the RGB image collection, resulting in a total of 62900 rgb images.<br> Regarding the acquisition of the data, the virtual camera was set to simulate a real camera (Astra RGBD camera) initially used in the CoLLaboratE task of a TV assembly. Thus, the rgb, and depth images (640x480) and the corresponding segmentation masks were extracted for all the animation frames (3700 different camera positions x 17 different backgrounds).</p>
A Dataset of Synthetic Images of Outdoor Scenes Taken from Sidewalks, for Temporal Semantic Segmentation Applications
<p>This dataset has been generated using the CARLA simulator (release 0.9.11), an open-source 3D simulator for experiments in autonomous vehicle, based on the Unreal Engine game engine. It comes with pre-made city environment maps. CARLA is distributed with several integrated maps as well as parameters to increase the variety in the dataset. In the release that we have used, there are 13 semantic segmentation classes: None, Building, Fence, Other, Pedestrian, Pole, Lane-marking, Road, Sidewalk, Vegetation, Vehicle, Wall, and Traffic sign. The "None" category corresponds to textures that are not part of an object, such as lawns which are not part of "Vegetation", or sky. In the “Other” category are found objects that are not included in the other classes like plant and flower pots. For smart mobility applications, the “Sidewalks” and “Road” classes are of particular importance to find the way forward, as well as “Buildings” and “Poles” for obstacle avoidance. Sequences are made of 4 images. The dataset is composed of 46436 frames (11609 sequences) partitioned in 41024 frames (10256 sequences) for train, 2696 frames (674 sequences) for validation, and 2716 for test (679 sequences). The size of the images is 800 x 600 (resp. width x height).</p> <p>Additionaly, we have generated another smaller dataset with images taken from 2 different viewpoints: one located on the road and the other located on the sidewalk. The number of frames for train/validation/test is respectively 7288 (1822 sequences) partitioned in 6344 (1687 sequences) for train, 416 frames (104 sequences) for validation, and 424 for test (106 sequences). This smaller dataset is aimed at showing the importance of the viewpoint in the result of semantic segmentation. This can be done by cross-validation: learning on images taken from a viewpoint located on the road and test on images with a viewpoint located on the sidewalk, and vice versa.</p>
A Large-scale Synthetic Pathological Dataset for Deep Learning-enabled Segmentation of Breast Cancer
<p>Dataset access for the paper: A Large-scale Synthetic Pathological Dataset for Deep Learning-enabled Segmentation of Breast Cancer</p>
Synthetic river flow videos dataset
<p>###### ######### ########## ###### ######### ##########<br> <strong> Synthetic river flow videos for evaluating image-based velocimetry methods</strong><br> ###### ######### ########## ###### ######### ##########<br> # Year : 2022<br> # Authors : G.Bodart (guillaume.bodart@inrae.fr), J.Le Coz (jerome.lecoz@inrae.fr), M.Jodeau (magali.jodeau@edf.fr), A.Hauet (alexandre.hauet@edf.fr)</p> <p>###<br> <em>This file describes the data attached to the article</em></p> <p>### -><strong> 00_article_cases</strong><br> ########################<br> <em>This folder contains the data used in the case studies: synthetic videos + reference files.</em></p> <p> - 00_reference_velocities<br> -> Reference velocities interpolated on a regular grid. Data are given in conventionnal units, i.e. m/s and m.</p> <p> - 01_XX<br> -> Data of the first case study</p> <p> - 02_XX<br> -> Data of the second case study</p> <p>### -> <strong>01_dev</strong><br> #############<br> <em>This folder contains the Python libraries and Mantaflow modified source code used in the paper. The libraries are provided as is. Feel free to contact us for support or guidelines.</em></p> <p> - lspiv<br> -> Python library used to extract, process and display results of LSPIV analysis carried out with Fudaa-LSPIV</p> <p> - mantaflow-modified<br> -> Modified version of Mantaflow described in the article. Installation instructions can be found at http://mantaflow.com</p> <p> - syri<br> -> Python library used to extract, process and display fluid simulations carried out on Mantaflow and Blender. (Require the lspiv library)</p> <p>### -> <strong>02_dataset</strong><br> #################<br> <em>This folder contains synthetic videos generated with the method described in the article. The fluid simulation parameters, and thus the reference velocities, are the same as those presented in the article. </em></p> <p> - The videos can be used freely. Please consider citing the corresponding paper.</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.