Dataset Description

OpenImage-O is an OOD benchmark subset derived from Open Images dataset. This split is derived from the OpenOOD benchmark OOD evaluation splits.

Homepage: https://storage.googleapis.com/openimages/web/index.html OpenOOD Benchmark: https://github.com/Jingkang50/OpenOOD/

  Citation

@article{kuznetsova2020open, title={The open images dataset v4: Unified image classification, object detection, and visual relationship detection at scale}… See the full description on the dataset page: https://huggingface.co/datasets/torch-uncertainty/Openimage-O.

https://i.imgur.com/ztezlER.png" alt="Image example">

Overview

This dataset contains 627 images of various vehicle classes for object detection. These images are derived from the Open Images open source computer vision datasets.

This dataset only scratches the surface of the Open Images dataset for vehicles!

https://i.imgur.com/4ZHN8kk.png" alt="Image example">

Use Cases

• Train object detector to differentiate between a car, bus, motorcycle, ambulance, and truck.
• Checkpoint object detector for autonomous vehicle detector
• Test object detector on high density of ambulances in vehicles
• Train ambulance detector
• Explore the quality and range of Open Image dataset

Tools Used to Derive Dataset

https://i.imgur.com/1U0M573.png" alt="Image example">

These images were gathered via the OIDv4 Toolkit This toolkit allows you to pick an object class and retrieve a set number of images from that class with bound box lables.

We provide this dataset as an example of the ability to query the OID for a given subdomain. This dataset can easily be scaled up - please reach out to us if that interests you.

Context

The dataset is released as CSV files. Each line in a CSV file corresponds to one data sample, which consists of images and annotations that indicate whether two faces in the photo are looking at each other.

Content

Each image is specified using an image ID/url and two face bounding boxes (top-left and bottom-right coordinates).

The Image URL serves as a preview of the image. Please access the image from OpenImageV4 using Image ID if the original image is removed from the public domain.

Each annotation is a boolean from the set {0, 1}. A value of 1 (0) means both faces are (not) looking at each other.

Each line in the CSV files has the following entries: * ImageID (string): The image ID in OpenImageV4. * ImageUrl (string): Url of image. * Annotation (boolean): Human annotation that indicates if both faces are looking at each other. * XminBoxA (float): Top-left column of the face bounding box A normalized with respect to width. * YminBoxA (float): Top-left row of the face bounding box A normalized with respect to height. * XmaxBoxA (float): Bottom-right column of the face bounding box A normalized with respect to width. * YmaxBoxA (float): Bottom-right row of the face bounding box A normalized with respect to height. * XminBoxB (float): Top-left column of the face bounding box B normalized with respect to width. * YminBoxB (float): Top-left row of the face bounding box B normalized with respect to height. * XmaxBoxB (float): Bottom-right column of the face bounding box B normalized with respect to width. * YmaxBoxB (float): Bottom-right row of the face bounding box B normalized with respect to height.

Acknowledgements

Big Shoutout to Authors Ching-Hui Chen Raviteja Vemulapalli Yukun Zhu GitHub

Amazon Bin Image Dataset

The Amazon Bin Image Dataset contains 536,434 images and metadata from bins of a pod in an operating Amazon Fulfillment Center. The bin images in this dataset are captured as robot units carry pods as part of normal Amazon Fulfillment Center operations. This dataset has many images and the corresponding medadata.

The image files have three groups according to its naming scheme.

• A file name with 1~4 digits (1,200): 1.jpg ~ 1200.jpg
• A file name with 5 digits (99,999): 00001.jpg ~ 99999.jpg
• A file name with 6 digits (435,235): 100000.jpg ~ 535234.jpg

Amazon Bin Image Dataset File List dataset aims to provide a CSV file to contain all file locations and the quantity to help the analysis and distributed learning.

Documentation

• https://github.com/awslabs/open-data-docs/tree/main/docs/aft-vbi-pds

Download

• https://registry.opendata.aws/amazon-bin-imagery/
• https://github.com/awslabs/open-data-registry/blob/main/datasets/amazon-bin-imagery.yaml

General

For more details and the most up-to-date information please consult our project page: https://kainmueller-lab.github.io/fisbe.

Summary

• A new dataset for neuron instance segmentation in 3d multicolor light microscopy data of fruit fly brains
  • 30 completely labeled (segmented) images
  • 71 partly labeled images
  • 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)
• To the best of our knowledge, the first real-world benchmark dataset for instance segmentation of long thin filamentous objects
• A set of metrics and a novel ranking score for respective meaningful method benchmarking
• An evaluation of three baseline methods in terms of the above metrics and score

Abstract

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.

Dataset documentation:

We provide a detailed documentation of our dataset, following the Datasheet for Datasets questionnaire:

>> FISBe Datasheet

Our dataset originates from the FlyLight project, where the authors released a large image collection of nervous systems of ~74,000 flies, available for download under CC BY 4.0 license.

Files

• fisbe_v1.0_{completely,partly}.zip
  • contains the image and ground truth segmentation data; there is one zarr file per sample, see below for more information on how to access zarr files.
• fisbe_v1.0_mips.zip
  • maximum intensity projections of all samples, for convenience.
• sample_list_per_split.txt
  • a simple list of all samples and the subset they are in, for convenience.
• view_data.py
  • a simple python script to visualize samples, see below for more information on how to use it.
• dim_neurons_val_and_test_sets.json
  • a list of instance ids per sample that are considered to be of low intensity/dim; can be used for extended evaluation.
• Readme.md
  • general information

How to work with the image files

Each sample consists of a single 3d MCFO image of neurons of the fruit fly.
For each image, we provide a pixel-wise instance segmentation for all separable neurons.
Each sample is stored as a separate zarr file (zarr is a file storage format for chunked, compressed, N-dimensional arrays based on an open-source specification.").
The image data ("raw") and the segmentation ("gt_instances") are stored as two arrays within a single zarr file.
The segmentation mask for each neuron is stored in a separate channel.
The order of dimensions is CZYX.

We recommend to work in a virtual environment, e.g., by using conda:

conda create -y -n flylight-env -c conda-forge python=3.9
conda activate flylight-env

How to open zarr files

1. Install the python zarr package:

   pip install zarr

2. Opened a zarr file with:
   

   import zarr
raw = zarr.open(
seg = zarr.open(

   # optional:
import numpy as np
raw_np = np.array(raw)

Zarr arrays are read lazily on-demand.
Many functions that expect numpy arrays also work with zarr arrays.
Optionally, the arrays can also explicitly be converted to numpy arrays.

How to view zarr image files

We recommend to use napari to view the image data.

1. Install napari:

   pip install "napari[all]"

2. Save the following Python script:
   

   import zarr, sys, napari

   raw = zarr.load(sys.argv[1], mode='r', path="volumes/raw")
gts = zarr.load(sys.argv[1], mode='r', path="volumes/gt_instances")

   viewer = napari.Viewer(ndisplay=3)
for idx, gt in enumerate(gts):
viewer.add_labels(
gt, rendering='translucent', blending='additive', name=f'gt_{idx}')
viewer.add_image(raw[0], colormap="red", name='raw_r', blending='additive')
viewer.add_image(raw[1], colormap="green", name='raw_g', blending='additive')
viewer.add_image(raw[2], colormap="blue", name='raw_b', blending='additive')
napari.run()

3. Execute:

   python view_data.py 

Metrics

• S: Average of avF1 and C
• avF1: Average F1 Score
• C: Average ground truth coverage
• clDice_TP: Average true positives clDice
• FS: Number of false splits
• FM: Number of false merges
• tp: Relative number of true positives

For more information on our selected metrics and formal definitions please see our paper.

Baseline

To showcase the FISBe dataset together with our selection of metrics, we provide evaluation results for three baseline methods, namely PatchPerPix (ppp), Flood Filling Networks (FFN) and a non-learnt application-specific color clustering from Duan et al..
For detailed information on the methods and the quantitative results please see our paper.

License

The FlyLight Instance Segmentation Benchmark (FISBe) dataset is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.

Citation

If you use FISBe in your research, please use the following BibTeX entry:

@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}
}

Acknowledgments

We thank Aljoscha Nern for providing unpublished MCFO images as well as Geoffrey W. Meissner and the entire FlyLight Project Team for valuable
discussions.
P.H., L.M. and D.K. were supported by the HHMI Janelia Visiting Scientist Program.
This work was co-funded by Helmholtz Imaging.

Changelog

There have been no changes to the dataset so far.
All future change will be listed on the changelog page.

Contributing

If you would like to contribute, have encountered any issues or have any suggestions, please open an issue for the FISBe dataset in the accompanying github repository.

All contributions are welcome!

https://i.imgur.com/4LoSPQh.png" alt="Image example">

Overview

This dataset contains 581 images of various shellfish classes for object detection. These images are derived from the Open Images open source computer vision datasets.

This dataset only scratches the surface of the Open Images dataset for shellfish!

https://i.imgur.com/oMK91v6.png" alt="Image example">

Use Cases

• Train object detector to differentiate between a lobster, shrimp, and crab.
• Train object dector to differentiate between shellfish
• Object detection dataset across different sub-species
• Object detection among related species
• Test object detector on highly related objects
• Train shellfish detector
• Explore the quality and range of Open Image dataset

Tools Used to Derive Dataset

https://i.imgur.com/1U0M573.png" alt="Image example">

These images were gathered via the OIDv4 Toolkit This toolkit allows you to pick an object class and retrieve a set number of images from that class with bound box lables.

We provide this dataset as an example of the ability to query the OID for a given subdomain. This dataset can easily be scaled up - please reach out to us if that interests you.

Context

Labeled datasets are useful in machine learning research.

Content

This public dataset contains approximately 9 million URLs and metadata for images that have been annotated with labels spanning more than 6,000 categories.

Tables: 1) annotations_bbox 2) dict 3) images 4) labels

Update Frequency: Quarterly

Querying BigQuery Tables

Fork this kernel to get started.

Acknowledgements

https://bigquery.cloud.google.com/dataset/bigquery-public-data:open_images

https://cloud.google.com/bigquery/public-data/openimages

APA-style citation: Google Research (2016). The Open Images dataset [Image urls and labels]. Available from github: https://github.com/openimages/dataset.

Use: The annotations are licensed by Google Inc. under CC BY 4.0 license.

The images referenced in the dataset are listed as having a CC BY 2.0 license. Note: while we tried to identify images that are licensed under a Creative Commons Attribution license, we make no representations or warranties regarding the license status of each image and you should verify the license for each image yourself.

Banner Photo by Mattias Diesel from Unsplash.

Inspiration

Which labels are in the dataset? Which labels have "bus" in their display names? How many images of a trolleybus are in the dataset? What are some landing pages of images with a trolleybus? Which images with cherries are in the training set?

Fake Image Dataset

Fake Image Dataset is now open-sourced at huggingface (InfImagine Organization) and openxlab. ↗ It consists of two folders, ImageData and MetaData. ImageData contains the compressed packages of the Fake Image Dataset, while MetaData contains the labeling information of the corresponding data indicating whether they are real or fake. Sentry-Image is now open-sourced at Sentry-Image (github repository) which provides the SOTA fake image detection models in… See the full description on the dataset page: https://huggingface.co/datasets/InfImagine/FakeImageDataset.

Collection of truck images, from a side point view, used to extract information about truck axles, collected on a highway in the State of São Paulo, Brazil. This is still a work in progress dataset and will be updated regularly, as new images are acquired. More info can be found on: Researchgate Lab Page, OrcID Profiles, or ITS Lab page on Github.

The dataset includes 725 cropped images of trucks, taken with three different cameras, on five different locations.

• 725 images
• Format: JPG
• Resolution: 1920xVarious, 96dpi, 24bits
• Naming pattern:

If this dataset helps in any way your research, please feel free to contact the authors. We really enjoy knowing about other researcher's projects and how everybody is making use of the images on this dataset. We are also open for collaborations and to answer any questions. We also have a paper that uses this dataset, so if you want to officially cite us in your research, please do so! We appreciate it!

Marcomini, Leandro Arab, and André Luiz Cunha. "Truck Axle Detection with Convolutional Neural Networks." arXiv preprint arXiv:2204.01868 (2022).

This dataset contains scanning electron microscope (SEM) images and labels from our paper "Towards Unsupervised SEM Image Segmentation for IC Layout Extraction", which are licensed under a Creative Commons Attribution 4.0 International License (CC-BY 4.0). The SEM images cover the logic area of the metal-1 (M1) and metal-2 (M2) layers of a commercial IC produced on a 128 nm technology node. We used an electron energy of 15 keV with a backscattered electron detector and a dwell time of 3 μs for SEM capture. The images are 4096×3536 pixels in size, with a resolution of 14.65 nm per pixel and 10% overlap. We discarded images on the logic area boundaries and publish the remaining ones in random order. We additionally provide labels for tracks and vias on the M2 layer, which are included as .svg files. For labeling, we employed automatic techniques, such as thresholding, edge detection, and size, position, and complexity filtering, before manually validating and correcting the generated labels. The labels may contain duplicates for detected vias. Tracks spanning multiple images may not be present in the label file of each image. The implementation of our approach, as well as accompanying evaluation and utility routines can be found in the following GitHub repository: https://github.com/emsec/unsupervised-ic-sem-segmentation Please make sure to always cite our study when using any part of our data set or code for your own research publications! @inproceedings {2023rothaug, author = {Rothaug, Nils and Klix, Simon and Auth, Nicole and B"ocker, Sinan and Puschner, Endres and Becker, Steffen and Paar, Christof}, title = {Towards Unsupervised SEM Image Segmentation for IC Layout Extraction}, booktitle = {Proceedings of the 2023 Workshop on Attacks and Solutions in Hardware Security}, series = {ASHES'23}, year = {2023}, month = {november}, keywords = {ic-layout-extraction;sem-image-segmentation;unsupervised-deep-learning;open-source-dataset}, url = {https://doi.org/10.1145/3605769.3624000}, doi = {10.1145/3605769.3624000}, isbn = {9798400702624}, publisher = {Association for Computing Machinery}, address = {New York, NY, USA} }

This dataset contains images used in the Open-source AI cookbook: https://github.com/huggingface/cookbook. Please make sure you optimize the assets before uploading them (e.g. using https://tinypng.com/).

This data repository for MedMNIST v1 is out of date! Please check the latest version of MedMNIST v2.

Abstract

We present MedMNIST, a collection of 10 pre-processed medical open datasets. MedMNIST is standardized to perform classification tasks on lightweight 28x28 images, which requires no background knowledge. Covering the primary data modalities in medical image analysis, it is diverse on data scale (from 100 to 100,000) and tasks (binary/multi-class, ordinal regression and multi-label). MedMNIST could be used for educational purpose, rapid prototyping, multi-modal machine learning or AutoML in medical image analysis. Moreover, MedMNIST Classification Decathlon is designed to benchmark AutoML algorithms on all 10 datasets; We have compared several baseline methods, including open-source or commercial AutoML tools. The datasets, evaluation code and baseline methods for MedMNIST are publicly available at https://medmnist.github.io/.

Please note that this dataset is NOT intended for clinical use.

We recommend our official code to download, parse and use the MedMNIST dataset:

pip install medmnist

Citation and Licenses

If you find this project useful, please cite our ISBI'21 paper as: Jiancheng Yang, Rui Shi, Bingbing Ni. "MedMNIST Classification Decathlon: A Lightweight AutoML Benchmark for Medical Image Analysis," arXiv preprint arXiv:2010.14925, 2020.

or using bibtex: @article{medmnist, title={MedMNIST Classification Decathlon: A Lightweight AutoML Benchmark for Medical Image Analysis}, author={Yang, Jiancheng and Shi, Rui and Ni, Bingbing}, journal={arXiv preprint arXiv:2010.14925}, year={2020} }

Besides, please cite the corresponding paper if you use any subset of MedMNIST. Each subset uses the same license as that of the source dataset.

PathMNIST

Jakob Nikolas Kather, Johannes Krisam, et al., "Predicting survival from colorectal cancer histology slides using deep learning: A retrospective multicenter study," PLOS Medicine, vol. 16, no. 1, pp. 1–22, 01 2019.

License: CC BY 4.0

ChestMNIST

Xiaosong Wang, Yifan Peng, et al., "Chestx-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases," in CVPR, 2017, pp. 3462–3471.

License: CC0 1.0

DermaMNIST

Philipp Tschandl, Cliff Rosendahl, and Harald Kittler, "The ham10000 dataset, a large collection of multisource dermatoscopic images of common pigmented skin lesions," Scientific data, vol. 5, pp. 180161, 2018.

Noel Codella, Veronica Rotemberg, Philipp Tschandl, M. Emre Celebi, Stephen Dusza, David Gutman, Brian Helba, Aadi Kalloo, Konstantinos Liopyris, Michael Marchetti, Harald Kittler, and Allan Halpern: “Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)”, 2018; arXiv:1902.03368.

License: CC BY-NC 4.0

OCTMNIST/PneumoniaMNIST

Daniel S. Kermany, Michael Goldbaum, et al., "Identifying medical diagnoses and treatable diseases by image-based deep learning," Cell, vol. 172, no. 5, pp. 1122 – 1131.e9, 2018.

License: CC BY 4.0

RetinaMNIST

DeepDR Diabetic Retinopathy Image Dataset (DeepDRiD), "The 2nd diabetic retinopathy – grading and image quality estimation challenge," https://isbi.deepdr.org/data.html, 2020.

License: CC BY 4.0

BreastMNIST

Walid Al-Dhabyani, Mohammed Gomaa, Hussien Khaled, and Aly Fahmy, "Dataset of breast ultrasound images," Data in Brief, vol. 28, pp. 104863, 2020.

License: CC BY 4.0

OrganMNIST_{Axial,Coronal,Sagittal}

Patrick Bilic, Patrick Ferdinand Christ, et al., "The liver tumor segmentation benchmark (lits)," arXiv preprint arXiv:1901.04056, 2019.

Xuanang Xu, Fugen Zhou, et al., "Efficient multiple organ localization in ct image using 3d region proposal network," IEEE Transactions on Medical Imaging, vol. 38, no. 8, pp. 1885–1898, 2019.

License: CC BY 4.0

The dataset contains multi-modal data from over 75,000 open access and de-identified case reports, including metadata, clinical cases, image captions and more than 130,000 images. Images and clinical cases belong to different medical specialties, such as oncology, cardiology, surgery and pathology. The structure of the dataset allows to easily map images with their corresponding article metadata, clinical case, captions and image labels. Details of the data structure can be found in the file data_dictionary.csv.

Almost 100,000 patients and almost 400,000 medical doctors and researchers were involved in the creation of the articles included in this dataset. The citation data of each article can be found in the metadata.parquet file.

Refer to the examples showcased in this GitHub repository to understand how to optimize the use of this dataset.

For a detailed insight about the contents of this dataset, please refer to this data article published in Data In Brief.

This repository introduces a dataset of obverse and reverse images of 319 unique Schlage SC1 keys, labeled with each key's bitting code. We make our data accessible in an HDF5 format, through arrays aligned where the Nth index of each array represents the Nth key, with keys sorted ascending by bitting code: /bittings: Each keys 1-9 bitting code, recorded from shoulder through the tip of the key, uint8 of shape (319, 5). /obverse: Obverse image of each key, uint8 of shape (319, 512, 512, 3). /reverse: Reverse image of each key, uint8 of shape (319, 512, 512, 3).

Full dataset details available on GitHub https://github.com/alexxke/keynet

OIP

  Dataset Summary

This dataset is derived from open-image-preferences-v1-binarized for our UnifiedReward-7B training. For further details, please refer to the following resources:

📰 Paper: https://arxiv.org/pdf/2503.05236 🪐 Project Page: https://codegoat24.github.io/UnifiedReward/ 🤗 Model Collections: https://huggingface.co/collections/CodeGoat24/unifiedreward-models-67c3008148c3a380d15ac63a 🤗 Dataset Collections:… See the full description on the dataset page: https://huggingface.co/datasets/CodeGoat24/OIP.

Dataset described in the paper "Synthbuster: Towards Detection of Diffusion Model Generated Images" (Quentin Bammey, 2023, Open Journal of Signal Processing)

This dataset contains synthetic, AI-generated images from 9 different models:

• DALL·E 2
• DALL·E 3
• Adobe Firefly
• Midjourney v5
• Stable Diffusion 1.3
• Stable Diffusion 1.4
• Stable Diffusion 2
• Stable Diffusion XL
• Glide

1000 images were generated per model. The images are loosely based on raise-1k images (Dang-Nguyen, Duc-Tien, et al. "Raise: A raw images dataset for digital image forensics." Proceedings of the 6th ACM multimedia systems conference. 2015.). For each image of the raise-1k dataset, a description was generated using the Midjourney /describe function and CLIP interrogator (https://github.com/pharmapsychotic/clip-interrogator/). Each of these prompts was manually edited to produce results as photorealistic as possible and remove living persons and artists names.

In addition to this, parameters were randomly selected within reasonable values for methods requiring so.

The prompts and parameters used for each method can be found in the `prompts.csv` file.

This dataset can be used to evaluate AI-generated image detection methods. We recommend matching the generated images with the real Raise-1k images, to evaluate whether the methods can distinguish the two of them. Raise-1k images are not included in the dataset, they can be downloaded separately at (http://loki.disi.unitn.it/RAISE/download.html).

None of the images suffered degradations such as JPEG compression or resampling, which leaves room to add your own degradations to test robustness to various transformation in a controlled manner.

The Quick Draw Dataset is a collection of 50 million drawings across 345 categories, contributed by players of the game "Quick, Draw!". The drawings were captured as timestamped vectors, tagged with metadata including what the player was asked to draw and in which country the player was located.

Example drawings: https://raw.githubusercontent.com/googlecreativelab/quickdraw-dataset/master/preview.jpg" alt="preview">

Experiments in: https://github.com/uleroboticsgroup/optimized_network_detect_burr_breakage dataset.csv: list of images dataset_train_test.csv: list of images divided in training set and test set images: original images in jpg mask: mask of the workpiece in png

About this dataset

This dataset contains 6,000 example images generated with the process described in Roboflow's How to Create a Synthetic Dataset tutorial.

The images are composed of a background (randomly selected from Google's Open Images dataset) and a number of fruits (from Horea94's Fruit Classification Dataset) superimposed on top with a random orientation, scale, and color transformation. All images are 416x550 to simulate a smartphone aspect ratio.

To generate your own images, follow our tutorial or download the code.

Example: https://blog.roboflow.ai/content/images/2020/04/synthetic-fruit-examples.jpg" alt="Example Image">

• Open Images is a dataset of ~9 million URLs to images that have been annotated with labels spanning over 6000 categories. This page aims to provide the download instructions and mirror sites for Open Images Dataset. Please visit the project page for more details on the dataset.

• These images contain the complete subsets of images for which instance segmentation and visual relations are annotated. The images are split into train (1,743,042), validation (41,620), and test (125,436) sets. The train set is also used in the Open Images Challenge 2018 and 2019. The images are rescaled to have at most 1024 pixels on their longest side while preserving their original aspect-ratio. The total size is 561GB. The images can be directly downloaded into a local directory from the CVDF AWS S3 cloud storage bucket:

1. The tsv files for the train set, in 10 partitions:
2. The tsv file for the validation set:
3. The tsv file for the test set: Dataset link