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.
24
datasets available to search
ShareScore release 0.9.0
Dataset results
24 results for “instance segmentation”
FISBe: A real-world benchmark dataset for instance segmentation of long-range thin filamentous structures
<h2>General</h2> <p>For more details and the most up-to-date information please consult our project page: <a href="https://kainmueller-lab.github.io/fisbe" target="_blank" rel="noopener">https://kainmueller-lab.github.io/fisbe</a>.</p> <h2>Summary</h2> <ul> <li>A new dataset for neuron instance segmentation in 3d multicolor light microscopy data of fruit fly brains <ul> <li>30 completely labeled (segmented) images</li> <li>71 partly labeled images</li> <li>altogether comprising ∼600 expert-labeled neuron instances (labeling a single neuron takes between 30-60 min on average, yet a difficult one can take up to 4 hours)</li> </ul> </li> <li>To the best of our knowledge, the first real-world benchmark dataset for instance segmentation of long thin filamentous objects</li> <li>A set of metrics and a novel ranking score for respective meaningful method benchmarking</li> <li>An evaluation of three baseline methods in terms of the above metrics and score</li> </ul> <h2>Abstract</h2> <p>Instance segmentation of neurons in volumetric light microscopy images of nervous systems enables groundbreaking research in neuroscience by facilitating joint functional and morphological analyses of neural circuits at cellular resolution. Yet said multi-neuron light microscopy data exhibits extremely challenging properties for the task of instance segmentation: Individual neurons have long-ranging, thin filamentous and widely branching morphologies, multiple neurons are tightly inter-weaved, and partial volume effects, uneven illumination and noise inherent to light microscopy severely impede local disentangling as well as long-range tracing of individual neurons. These properties reflect a current key challenge in machine learning research, namely to effectively capture long-range dependencies in the data. While respective methodological research is buzzing, to date methods are typically benchmarked on synthetic datasets. To address this gap, we release the FlyLight Instance Segmentation Benchmark (FISBe) dataset, the first publicly available multi-neuron light microscopy dataset with pixel-wise annotations. In addition, we define a set of instance segmentation metrics for benchmarking that we designed to be meaningful with regard to downstream analyses. Lastly, we provide three baselines to kick off a competition that we envision to both advance the field of machine learning regarding methodology for capturing long-range data dependencies, and facilitate scientific discovery in basic neuroscience.</p> <h2>Dataset documentation:</h2> <p>We provide a detailed documentation of our dataset, following the <a href="https://arxiv.org/abs/1803.09010" target="_blank" rel="noopener">Datasheet for Datasets</a> questionnaire:</p> <p><em>>> <a href="https://kainmueller-lab.github.io/fisbe/datasheet" target="_blank" rel="noopener">FISBe Datasheet</a></em></p> <p>Our dataset originates from the <a href="https://www.janelia.org/project-team/flylight" target="_blank" rel="noopener">FlyLight project</a>, where the authors released a large image collection of nervous systems of ~74,000 flies, <a href="https://gen1mcfo.janelia.org/cgi-bin/gen1mcfo.cgi" target="_blank" rel="noopener">available for download</a> under CC BY 4.0 license.</p> <h2>Files</h2> <ul> <li>fisbe_v1.0_{completely,partly}.zip <ul> <li>contains the image and ground truth segmentation data; there is one <em>zarr</em> file per sample, see below for more information on how to access <em>zarr</em> files.</li> </ul> </li> <li>fisbe_v1.0_mips.zip <ul> <li>maximum intensity projections of all samples, for convenience.</li> </ul> </li> <li>sample_list_per_split.txt <ul> <li>a simple list of all samples and the subset they are in, for convenience.</li> </ul> </li> <li>view_data.py <ul> <li>a simple python script to visualize samples, see below for more information on how to use it.</li> </ul> </li> <li>dim_neurons_val_and_test_sets.json <ul> <li>a list of instance ids per sample that are considered to be of low intensity/dim; can be used for extended evaluation.</li> </ul> </li> <li>Readme.md <ul> <li>general information</li> </ul> </li> </ul> <h2>How to work with the image files</h2> <p>Each sample consists of a single 3d MCFO image of neurons of the fruit fly.<br>For each image, we provide a pixel-wise instance segmentation for all separable neurons.<br>Each sample is stored as a separate <em>zarr</em> file (<a href="https://zarr.readthedocs.io" target="_blank" rel="noopener">zarr</a> is a file storage format for chunked, compressed, N-dimensional arrays based on an open-source specification.").<br>The image data ("raw") and the segmentation ("gt_instances") are stored as two arrays within a single zarr file.<br>The segmentation mask for each neuron is stored in a separate channel.<br>The order of dimensions is CZYX.</p> <p>We recommend to work in a virtual environment, e.g., by using conda:</p> <p><code>conda create -y -n flylight-env -c conda-forge python=3.9</code><br><code>conda activate flylight-env</code></p> <h3>How to open <em>zarr</em> files</h3> <ol> <li>Install the python zarr package: <pre><code>pip install zarr</code></pre> </li> <li>Opened a zarr file with:<br> <p><code>import zarr</code><br><code>raw = zarr.open(<path_to_zarr>, mode='r', path="volumes/raw")</code><br><code>seg = zarr.open(<path_to_zarr>, mode='r', path="volumes/gt_instances")</code></p> <p><code># optional:</code><br><code>import numpy as np</code><br><code>raw_np = np.array(raw)</code></p> </li> </ol> <p>Zarr arrays are read lazily on-demand.<br>Many functions that expect numpy arrays also work with zarr arrays.<br>Optionally, the arrays can also explicitly be converted to numpy arrays.</p> <h3>How to view <em>zarr</em> image files</h3> <p>We recommend to use <a href="https://napari.org" target="_blank" rel="noopener">napari</a> to view the image data.</p> <ol> <li>Install napari: <pre><code>pip install "napari[all]"</code></pre> </li> <li>Save the following Python script: <br> <p><code>import zarr, sys, napari</code></p> <p><code>raw = zarr.load(sys.argv[1], mode='r', path="volumes/raw")</code><br><code>gts = zarr.load(sys.argv[1], mode='r', path="volumes/gt_instances")</code></p> <p><code>viewer = napari.Viewer(ndisplay=3)</code><br><code>for idx, gt in enumerate(gts):</code><br><code> viewer.add_labels(</code><br><code> gt, rendering='translucent', blending='additive', name=f'gt_{idx}')</code><br><code>viewer.add_image(raw[0], colormap="red", name='raw_r', blending='additive')</code><br><code>viewer.add_image(raw[1], colormap="green", name='raw_g', blending='additive')</code><br><code>viewer.add_image(raw[2], colormap="blue", name='raw_b', blending='additive')</code><br><code>napari.run()</code></p> </li> <li>Execute: <pre><code>python view_data.py <path-to-file>/R9F03-20181030_62_B5.zarr</code></pre> </li> </ol> <h2>Metrics</h2> <ul> <li>S: Average of avF1 and C</li> <li>avF1: Average F1 Score</li> <li>C: Average ground truth coverage</li> <li>clDice_TP: Average true positives clDice</li> <li>FS: Number of false splits</li> <li>FM: Number of false merges</li> <li>tp: Relative number of true positives</li> </ul> <p>For more information on our selected metrics and formal definitions please see <a href="https://arxiv.org/abs/2404.00130" target="_blank" rel="noopener">our paper</a>.</p> <h2>Baseline</h2> <p>To showcase the FISBe dataset together with our selection of metrics, we provide evaluation results for three baseline methods, namely <a href="https://github.com/Kainmueller-Lab/PatchPerPix" target="_blank" rel="noopener">PatchPerPix (ppp)</a>, <a href="https://github.com/google/ffn" target="_blank" rel="noopener">Flood Filling Networks (FFN)</a> and a non-learnt application-specific <a href="https://www.biorxiv.org/content/10.1101/2020.06.07.138941v1" target="_blank" rel="noopener">color clustering from Duan et al.</a>.<br>For detailed information on the methods and the quantitative results please see <a href="https://arxiv.org/abs/2404.00130" target="_blank" rel="noopener">our paper</a>.</p> <h2>License</h2> <p>The FlyLight Instance Segmentation Benchmark (FISBe) dataset is licensed under the <a href="https://creativecommons.org/licenses/by/4.0" target="_blank" rel="noopener">Creative Commons Attribution 4.0 International (CC BY 4.0) license</a>.</p> <h2>Citation</h2> <p>If you use <em>FISBe</em> in your research, please use the following BibTeX entry: </p> <pre><code>@misc{mais2024fisbe, title = {FISBe: A real-world benchmark dataset for instance segmentation of long-range thin filamentous structures}, author = {Lisa Mais and Peter Hirsch and Claire Managan and Ramya Kandarpa and Josef Lorenz Rumberger and Annika Reinke and Lena Maier-Hein and Gudrun Ihrke and Dagmar Kainmueller}, year = 2024, eprint = {2404.00130}, archivePrefix ={arXiv}, primaryClass = {cs.CV} }</code></pre> <h2>Acknowledgments</h2> <p>We thank Aljoscha Nern for providing unpublished MCFO images as well as Geoffrey W. Meissner and the entire FlyLight Project Team for valuable<br>discussions.<br>P.H., L.M. and D.K. were supported by the HHMI Janelia Visiting Scientist Program.<br>This work was co-funded by Helmholtz Imaging.</p> <h2>Changelog</h2> <p>There have been no changes to the dataset so far.<br>All future change will be listed <a href="https://kainmueller-lab.github.io/fisbe/changelog" target="_blank" rel="noopener">on the changelog page</a>.</p> <h2>Contributing</h2> <p>If you would like to contribute, have encountered any issues or have any suggestions, please <a href="https://github.com/Kainmueller-Lab/fisbe/issues" target="_blank" rel="noopener">open an issue</a> for the FISBe dataset in the accompanying github repository.</p> <p>All contributions are welcome!</p>
Embrapa Wine Grape Instance Segmentation Dataset – Embrapa WGISD
<p><strong>Embrapa Wine Grape Instance Segmentation Dataset – Embrapa WGISD</strong></p> <p>For a detailed description of this dataset, following the <em>datasheet for the datasets</em> recommendation proposed by <a href="https://arxiv.org/abs/1803.09010">Gebru et al.</a>, check the <strong>README.md</strong> file.</p> <p><strong>Motivation for Dataset Creation</strong></p> <p><em>Why was the dataset created?</em></p> <p>Embrapa WGISD (<em>Wine Grape Instance Segmentation Dataset</em>) was created to provide images and annotation to study <em>object detection and instance segmentation</em> for image-based monitoring and field robotics in viticulture. It provides instances from five different grape varieties taken on field. These instances shows variance in grape pose,<br> illumination and focus, including genetic and phenological variations such as shape, color and compactness.</p> <p><em>What (other) tasks could the dataset be used for?</em></p> <p>Possible uses include relaxations of the instance segmentation problem: classification (Is a grape in the image?), semantic segmentation (What are the "grape pixels" in the image?), and object detection (Where are the grapes in the image?). The WGISD can also be used in grape variety identification.</p> <p> </p>
NPM3D dataset with instance labels used in paper "Toward Accurate Instance Segmentation in Large-scale LiDAR Point Clouds"
<p>NPM3D (https://npm3d.fr/paris-carla-3d) consists of mobile laser scanning (MLS) point clouds collected in four different regions in the French cities of Paris and Lille, where each point has been annotated with two labels: one that assigns it to one out of 10 semantic categories and another one that assigns it to an object instance. When inspecting the data, we found 9 cases where multiple tree instances had not been separated correctly (i.e., they had the same ground truth instance label). These cases were manually corrected using the CloudCompare software (https://www.cloudcompare.org), and 35 individual tree instances were obtained. Our variant of the dataset with 10 semantic categories and enhanced instance labels is publicly available.</p>
Cell Instance Segmentation Dataset
<p>This project contains the data for the paper:</p> <p>A. Bouyssoux, R. Fezzani and J. -C. Olivo-Marin, "<em>Cell Instance Segmentation Using Z-Stacks In Digital Cytopathology,</em>" 2022 IEEE International Symposium on Biomedical Imaging (ISBI), 2022.</p> <p>The code associated with this project is available at: <a href="https://gitlab.com/vitadx/articles/zstacks_cell_instance_segmentation">https://gitlab.com/vitadx/articles/zstacks_cell_instance_segmentation</a></p> <p>A new large Cell Instance Segmentation Dataset (CISD) is introduced. It comprises 3911 samples containing at least two touching or overlapping urothelial cells. Cell instances were manually annotated by trained cytotechnicians. All samples are extracted from 30 digital cytology slides stained with nine variations of Papanicolaou staining. The cytology slides are prepared from urine samples from healthy patients, using a Hologic ThinPrep®5000 processor, and routinely stained with the Agilent Dako CoverStainer®. The slides are finally digitized using a Hamamatsu NanoZoomer®S360 with 21 focal planes and centered on the best focus plane determined by the scanner autofocus.</p> <p>Note that cell instances with a bounding box width/height smaller than 10% of the sample width/height, as well as red blood cells and neutrophil cells were automatically filtered out because under-represented in the available data.</p> <p>Each sample is considered in three different manners in the CISD, allowing experimentation with different methods<br> for handling Z-stack data and comparison with simple 2D acquisition:</p> <ul> <li>Center slice: the best focus plane only as determined by the scanner, which is equivalent to a 2D slide acquisition.</li> <li>Extended Depth of Field (EDF): the 21 planes merged in an image where all textured parts appear in focus.</li> <li>Raw Z-stack: the volume composed of the 21 focal planes.</li> </ul> <p>Once extracted, the dataset folder contains three subfolders, one for each of the three types of samples described here above, containing the images and stack of images. A JSON file contains the instance masks, encoded in RLE format.</p>
Instance Segmentation of Dislocations in TEM Images
<p>This is the dataset and software for the IEEE publication <em>Instance Segmentation of Dislocations in TEM Images </em>to be published in 2023.</p>
FOR-instance: a UAV laser scanning benchmark dataset for semantic and instance segmentation of individual trees
<p>The challenge of accurately segmenting individual trees from laser scanning data hinders the assessment of crucial tree parameters necessary for effective forest management, impacting many downstream applications. While dense laser scanning offers detailed 3D representations, automating the segmentation of trees and their structures from point clouds remains difficult. The lack of suitable benchmark datasets and reliance on small datasets have limited method development. The emergence of deep learning models exacerbates the need for standardized benchmarks. Addressing these gaps, the FOR-instance data represent a novel benchmarking dataset to enhance forest measurement using dense airborne laser scanning data, aiding researchers in advancing segmentation methods for forested 3D scenes.</p> <p>In this repository, users will find forest laser scanning point clouds from unamnned aerial vehicle (using Riegl sensors) that are manually segmented according to the individual trees (1130 trees) and semantic classes. The point clouds are subdivided into five data collections representing different forests in Norway, the Czech Republic, Austria, New Zealand, and Australia. </p> <p>These data are meant to be used either for developement of new methods (using the dev data) or for testing of exisitng methods (test data). The data splits are provided in the data_split_metadata.csv file.</p> <p>A full description of the FOR-instance data can be found at <a href="http://arxiv.org/abs/2309.01279">http://arxiv.org/abs/2309.01279</a> </p>
Robust semi-automatic vessel tracing in the human retinal image by an instance segmentation neural network
Open the record for dataset details and reuse information.
VTT-Coffee: A small dataset for instance segmentation and iterative learning
<p>VTT-Coffee is a small scale dataset containing microscopic images of coffee grinds and annotations of their contours (segmentation masks).</p> <p>The dataset is useful to benchmark instance segmentations methods, especially for use cases containing just one class with a lot of variability.</p>
3D nuclei instance segmentation dataset of fluorescence microscopy volumes of C. elegans
<p>The dataset consists of 28 confocal microscopy volumes of C. elegans worms at the L1 stage and corresponding stacks of densely annotated nuclei instance segmentation masks.</p> <p>* 28 raw images and corresponding masks of average dimension (xyz) 1050 x 140 x 140<br> * Pixelsize (xyz): 0.116 x 0.116 x 0.122μm<br> * Microscope: Leica confocal microscopy, 63x oil objective</p> <p><br> The original raw data and preliminary annotations were part of the following publication (please cite if you use the dataset):<br> <br> <em>Long, F., Peng, H., Liu, X., Kim, S. K., & Myers, E. (2009). A 3D digital atlas of C. elegans and its application to single-cell analyses. Nature methods, 6(9), 667-672.</em></p> <p>The nuclei annotation masks were further manually curated by Dagmar Kainmueller (MDC Berlin) for the following publication:</p> <p><em>Hirsch, P., & Kainmueller, D. (2020). An auxiliary task for learning nuclei segmentation in 3d microscopy images. In Medical Imaging with Deep Learning (pp. 304-321). PMLR.</em></p> <p>We provide the dataset already structured into the train/validation/test split as used by the above as well as the following publications: </p> <p><em>Weigert, M., Schmidt, U., Haase, R., Sugawara, K., & Myers, G. (2020). Star-convex polyhedra for 3d object detection and segmentation in microscopy. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (pp. 3666-3673).</em><br> </p> <p> </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>
Night and Day Instance Segmented Park (NDISPark) Dataset: a Collection of Images taken by Day and by Night for Vehicle Detection, Segmentation and Counting in Parking Areas
<p><strong>The Dataset</strong></p> <p>A collection of images of parking lots for <em>vehicle detection, segmentation, and counting</em>.<br> Each image is <em>manually</em> labeled with pixel-wise masks and bounding boxes localizing vehicle instances.<br> The dataset includes about 250 images depicting several parking areas describing most of the problematic situations that we can find in a real scenario: seven different cameras capture the images under various weather conditions and viewing angles. Another challenging aspect is the presence of partial occlusion patterns in many scenes such as obstacles (trees, lampposts, other cars) and shadowed cars.<br> The main peculiarity is that <em>images are taken during the day and the night</em>, showing utterly different lighting conditions.</p> <p>We suggest a three-way split (train-validation-test). The train split contains images taken during the daytime while validation and test splits include images gathered at night.<br> In line with these splits we provide some annotation files:</p> <ul> <li> <p><em>train_coco_annotations.json</em> and <em>val_coco_annotations.json</em> --> JSON files that follow the golden standard MS COCO data format (for more info see <a href="https://cocodataset.org/#format-data">https://cocodataset.org/#format-data</a>) for the training and the validation splits, respectively. All the vehicles are labeled with the COCO category<em> 'car'</em>. They are suitable for vehicle detection and instance segmentation.</p> </li> <li> <p><em>train_dot_annotations.csv</em> and <em>val_dot_annotations.csv</em> --> CSV files that contain xy coordinates of the centroids of the vehicles for the training and the validation splits, respectively. Dot annotation is commonly used for the visual counting task.</p> </li> <li> <p><em>ground_truth_test_counting.csv</em> --> CSV file that contains the number of vehicles present in each image. It is only suitable for testing vehicle counting solutions.</p> </li> </ul> <p> </p> <p><strong>Citing our work</strong></p> <p>If you found this dataset useful, please cite the following paper</p> <blockquote> <pre>@inproceedings{Ciampi_visapp_2021, doi = {10.5220/0010303401850195}, url = {https://doi.org/10.5220%2F0010303401850195}, year = 2021, publisher = {{SCITEPRESS} - Science and Technology Publications}, author = {Luca Ciampi and Carlos Santiago and Joao Costeira and Claudio Gennaro and Giuseppe Amato}, title = {Domain Adaptation for Traffic Density Estimation}, booktitle = {Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications} } </pre> </blockquote> <p>and this Zenodo Dataset</p> <blockquote> <pre>@dataset{ciampi_ndispark_6560823, author = {Luca Ciampi and Carlos Santiago and Joao Costeira and Claudio Gennaro and Giuseppe Amato}, title = {{Night and Day Instance Segmented Park (NDISPark) Dataset: a Collection of Images taken by Day and by Night for Vehicle Detection, Segmentation and Counting in Parking Areas}}, month = may, year = 2022, publisher = {Zenodo}, version = {1.0.0}, doi = {10.5281/zenodo.6560823}, url = {https://doi.org/10.5281/zenodo.6560823} } </pre> </blockquote> <p> </p> <p><strong>Contact Information</strong></p> <p>If you would like further information about the dataset or if you experience any issues downloading files, please contact us at <a href="mailto:mobdrone@isti.cnr.it">luca.ciampi@isti.cnr.it</a></p> <p> </p>
SOCRATES Plittersdorf Instance Segmentation Dataset
<p>A subset of video frames captured by the SOCRATES stereo camera trap in a wildlife park in Bonn, Germany between February and July of 2022, with corresponding instance segmentation annotations in the COCO format.</p>
2D Cell Instance Segmentation Dataset of Master Thesis "Enhancing Cell Instance Segmentation in 3D Microscopy using Self-Supervised ViTs"
<p>This is the 2D cell instance segmentation dataset of master thesis "Enhancing Cell Instance Segmentation in 3D Microscopy using Self-Supervised ViTs". The data is originally from the BBBC038 dataset for the Kaggle 2018 Data Science Bowl. We keep the images with annotations and the final dataset comprises 670 images in the training set and 171 images in the test set. We also converted the dataset to MSCOCO format for convenience.</p>
3D Cell Instance Segmentation Dataset of Master Thesis "Enhancing Cell Instance Segmentation in 3D Microscopy using Self-Supervised ViTs"
<p>This is the 3D cell instance segmentation dataset of master thesis "Enhancing Cell Instance Segmentation in 3D Microscopy using Self-Supervised ViTs". The data is originally from the BBBC027 dataset comprising 30 sets of 3D image sets with high SNR level of quality. Because only the 4 to 100 slides of every 3D image set are annotated and the main difference between annotated and unannoated slides are the brightness, to avoiding misunderstanding by the algorithms, we removed the unannotated slides and separate every slide as individual images. We also converted the dataset to MSCOCO format for convenience.</p>
ReSyRIS: Real-Synthetic Rock Instance Segmentation dataset
<pre># ReSyRIS The Real-Synthetic Rock Instance Segmentation dataset (ReSyRIS) is created for training and evaluation of rock segmentation, detection and instance segmentation in (quasi-)extra-terrestrial environments. It consists of a set of annotated, real images of rocks on a lunar-like surface, a precisely mimicked synthetic version thereof, and respective synthetic assets for training data generation. In the folders, you find the following structure: - `stone_models`: all 36 .obj files of the 3d reconstructed stones - `test_data_realworld`: the real world recordings with accompanying ground truth - `test_data_synthetic`: the synthetic renderings matching approximately the real world recordings, with accompanying ground truth - `oaisys`: config files for rendering synthetic training data with oaisys If you find this dataset useful for your work please consider citing our paper: https://elib.dlr.de/194113/. </pre>
NPM3D dataset with instance label. Dataset used in paper "A Review of Panoptic Segmentation for Mobile Mapping Point Clouds"
<p>NPM3D is a public benchmark for point cloud semantic segmentation, with 10 classes including: ground, building, pole (road sign and traffic light), bollard, trash can, barrier, pedestrian, car, natural (vegetation) and unclassified. Results are evaluated only w.r.t. 9 classes, disregarding the "unclassified" label. The data has been captured with a mapping-grade mobile laser scanning system in different cities in France. There are 4 regions designated for training, all captured in Paris and Lille; and 3 regions for testing, captured in Dijon and Ajaccio. The standard 10-class version described above has actually been derived from a more fine-grained version of the dataset by keeping only the most frequent labels. The original annotations feature 50 different semantic classes (most of which are very rare), and also individual object instance labels for the training regions. For panoptic segmentation, a new version has been generated that still uses the 10 semantic category labels listed above, but also includes instance labels. The classes ground, building and barrier are considered "stuff" and are not separated into instances. As no instance labels are available for the 3 test regions, our version for panoptic (or pure instance) segmentation only contains 4 different regions from Paris and Lille. Instead of a fixed training/test split all experiments therefore use 4-fold cross-validation.</p>
Fraunhofer EZRT XXL-CT Instance Segmentation Me163
<p>The ’Me 163’ was a Second World War fighter airplane and a result of the German air force secret developments. One of these airplanes is currently owned and displayed in the historic aircraft exhibition of the ’Deutsches Museum’ in Munich, Germany. To gain insights with respect to its history, design and state of preservation, a complete CT scan was obtained using an industrial XXL-computer tomography scanner.<br><br>Using the CT data from the Me 163, all its details can visually be examined at various levels, ranging from the complete hull down to single sprockets and rivets. However, while a trained human observer can identify and interpret the volumetric data with all its parts and connections, a virtual dissection of the airplane and all its different parts would be quite desirable. Nevertheless, this means, that an instance segmentation of all components and objects of interest into disjoint entities from the CT data is necessary.</p> <p>As of currently, no adequate computer-assisted tools for automated or semi-automated segmentation<br>of such XXL-airplane data are available, in a first step, an interactive data annotation and object labeling process has been established. So far, seven sub-volumes from the Me 163 airplane have been annotated and labeled, whose results can potentially be used for various new applications in the field of digital heritage, non-destructive testing, or machine-learning. These annotated and labeled data sets are available here.</p>
MaskUKF: An Instance Segmentation Aided Unscented Kalman Filter for 6D Object Pose and Velocity Tracking - Accompanying Data
<p>Dataset and evaluation data associated to the publication "MaskUKF: An Instance Segmentation Aided Unscented Kalman Filter for 6D Object Pose and Velocity Tracking".</p>
Supplemental Data: Rapid Automated Mapping of Clouds on Titan with Instance Segmentation
<p>Supplemental data provided in conjuction with paper submitted to AGU JGR: Machine Learning and Computation.</p>
DLO Instance Segmentation dataset generated by Blender
<p>This dataset is released as part of the publication: "Adapter to facilitate Foundation Model Communication for DLO Instance Segmentation" - https://openreview.net/forum?id=EYBtzTQvdW</p> <pre><code><a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/"><img alt="Creative Commons License" style="border-width:0" src="https://i.creativecommons.org/l/by-sa/4.0/80x15.png" /></a><br />This work is licensed under a <a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/">Creative Commons Attribution-ShareAlike 4.0 International License</a>.</code></pre>
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.