19,997 machine learning datasets
19,997 dataset results
Accidental Turntables contains a challenging set of 41,212 images of cars in cluttered backgrounds, motion blur and illumination changes that serves as a benchmark for 3D pose estimation.
Scan Entities in 3D (ScanEnts3D) is a large-scale dataset which provides explicit correspondences between 369k objects across 84k natural referentural sentences, covering 705 real-world scenes.
This dataset is the image stimulus pool of 50 deepfake and 50 real images, used for the experiment in the study titled "Testing Human Ability To Detect Deepfake Images of Human Faces".
A large dataset of natural language descriptions for physical 3D objects in the ShapeNet dataset.
THU-FVFDT is a dataset containing raw finger vein and finger dorsal texture images of 220 different subjects. Images are captured in two different sessions with interval of about dozens of seconds. One session is for training and the other for testing. Four finger vein images and four finger dorsal texture images are captured simultaneously in each session. We only offer one of four images in that there is approximately no difference between them. The size of raw images is 720×576 pixels.
Dataset of low fidelity resolutions of the RANS equations over airfoils.
MiST (Modals In Scientific Text) is a dataset containing 3737 modal instances in five scientific domains annotated for their semantic, pragmatic, or rhetorical function.
It consists of 36 sequences, recorded in mines, tunnels, and other dark environments, totaling more than 145 minutes of stereo camera video and IMU data. In each sequence, the scene is illuminated by an onboard light of approximately 1350, 4500, or 9000 lumens. We accommodate both direct and indirect VIO methods by providing the geometric and photometric camera calibrations. The full dataset, including sensor data, calibration sequences, and evaluation scripts can be downloaded here.
The dataset contains 32 sequences for the evaluation of VI motion estimation methods, totalling ∼80 min of data. The dataset covers challenging conditions (mainly illumination changes and low textured environments) in different degrees and a wide rage of scenarios (including corridors, parking, classrooms, halls, etc.) from two different buildings at the University of Malaga. In general, we provide at least two different sequences within the same scenario, with different illumination conditions or following different trajectories. All sequences were recorded with our VI sensor handheld, except a few that were recorded while mounted in a car.
The Rotterdam EyePACS AIROGS dataset (in full, so including train and test) contains 113,893 color fundus images from 60,357 subjects and approximately 500 different sites with a heterogeneous ethnicity.
Histological images of colorectal cancer, derived from the TCGA database
FreCDo is a corpus for French dialect identification comprising 413,522 French text samples collected from public news websites in Belgium, Canada, France and Switzerland.
To encourage reproducible research, a labeled MultiRAW dataset containing>7k RAW images acquired using multiple camera sensors is made publicly accessible for RAW-domain processing.
Robust Summarization Evaluation Benchmark is a large human evaluation dataset consisting of over 22k summary-level annotations over state-of-the-art systems on three datasets.
This dataset contains the raw images for the dataset of Thermal Bridges on Building Rooftops (TBBR) dataset.
Early detection of retinal diseases is one of the most important means of preventing partial or permanent blindness in patients. One of the major stumbling blocks for manual retinal examination is the lack of a sufficient number of qualified medical personnel per capita to diagnose diseases. Computer-aided diagnosis systems (CAD) have proven to be very effective in helping physicians reduce the time taken to make a diagnosis and minimize variability in image interpretation. Still, they are not flexible enough to accommodate the simultaneous presence of multiple retinal diseases, which is a common situation in real-world applications. In the past years, few datasets that focus on the classification of numerous retinal pathologies present at the same time, i.e., multi-label classification have been proposed, but there are some shared problems with all of them, such as a narrow range of pathologies to classify, high level of class imbalance, low amount of samples for the underrepresented
FETA benchmark focuses on text-to-image and image-to-text retrieval in public car manuals and sales catalogue brochures. The FETA Car-Manuals dataset consists of a total of 349 PDF documents from 5 car manufacturers, namely Nissan, Toyota, Mazda, Renault, Chevrolet.
FETA benchmark focuses on text-to-image and image-to-text retrieval in public car manuals and sales catalogue brochures. The FETA IKEA dataset contains 26 documents with 7366 pages total, approximately 9574 images and 23927 texts automatically extracted from those pages.
Verifee is a dataset of news articles with fine-grained trustworthiness annotations. It contains over 10, 000 unique articles from almost 60 Czech online news sources. These are categorized into one of the 4 classes across the credibility spectrum we propose, raging from entirely trustworthy articles all the way to the manipulative ones.
Werewolf Among Us is a dataset multimodal dataset for modeling persuasion behaviors. It contains 199 dialogue transcriptions and videos captured in a multi-player social deduction game setting, 26,647 utterance level annotations of persuasion strategy, and game level annotations of deduction game outcomes.