19,997 machine learning datasets
19,997 dataset results
OmniCity is a dataset for omnipotent city understanding from multi-level and multi-view images. It contains multi-view satellite images as well as street-level panorama and mono-view images, constituting over 100K pixel-wise annotated images that are well-aligned and collected from 25K geo-locations in New York City. This dataset introduces a new task of fine-grained building instance segmentation on street-level panorama images. It also provides new problem settings for existing tasks, such as cross-view image matching, synthesis, segmentation, detection, etc., and facilitates the developing of new methods for large-scale city understanding, reconstruction, and simulation.
Training data for Hebrew morphological word segmentation
We introduce KPI-EDGAR, a novel dataset for Joint Named Entity Recognition and Relation Extraction building on financial reports uploaded to the Electronic Data Gathering, Analysis, and Retrieval (EDGAR) system, where the main objective is to extract Key Performance Indicators (KPIs) from financial documents (the named entity recognition part) and link them to their numerical values (the relation extraction part).
PersonPath22 is a large-scale multi-person tracking dataset containing 236 videos captured mostly from static-mounted cameras, collected from sources where we were given the rights to redistribute the content and participants have given explicit consent. Each video has ground-truth annotations including both bounding boxes and tracklet-ids for all the persons in each frame.
Occluded COCO is automatically generated subset of COCO val dataset, collecting partially occluded objects for a large variety of categories in real images in a scalable manner, where target object is partially occluded but the segmentation mask is connected.
Our dataset comprises $23.468$ non-labelled and $356$ labelled samples where each sample is $512 \times 512 \times 1$ dimensional IR image collected with the thermographic measurement specifications. Some samples contain scars, shadows, salt \& pepper noises and contrast burst regions, demonstrating that realistic laminar-turbulent flow observation scenarios are subject to high noise. Besides, a laminar flow area may occur brighter or darker as compared to the regions in a turbulent flow. Due to some effect (e.g. shadowing the sun) it is even possible that, in one part of the image, the laminar flow area appears darker, and in another part, it appears brighter than the turbulent flow area.
This paper is a condensed report on the second year of the Touché shared task on argument retrieval held at CLEF 2021. With the goal to provide a collaborative platform for researchers, we organized two tasks: (1) supporting individuals in finding arguments on controversial topics of social importance and (2) supporting individuals with arguments in personal everyday comparison situations.
WDC SOTAB is a benchmark that features two annotation tasks: Column Type Annotation and Columns Property Annotation. The goal of the Column Type Annotation (CTA) task is to annotate the columns of a table with 91 Schema.org types, such as telephone, duration, Place, or Organization. The goal of the Columns Property Annotation (CPA) task is to annotate pairs of table columns with one out of 176 Schema.org properties, such as gtin13, startDate, priceValidUntil, or recipeIngredient. The benchmark consists of 59,548 tables annotated for CTA and 48,379 tables annotated for CPA originating from 74,215 different websites. The tables are split into training-, validation- and test sets for both tasks. The tables cover 17 popular Schema.org types including Product, LocalBusiness, Event, and JobPosting. The tables originate from the Schema.org Table Corpus.
Electrophysiological data from implanted electrodes in the human brain are rare, and therefore scientific access to it has remained somewhat exclusive. Here we present a freely-available curated library of implanted electrocorticographic (ECoG) data and analyses for 16 benchmark behavioral experiments, with 204 individual datasets from 34 patients made with the same amplifiers (at the same sampling rate and filter settings). In every case, electrode positions have been carefully registered to brain anatomy. A large set of fully-commented analysis scripts to interpret these data using modern techniques is embedded in the library alongside the data. All data, anatomic correlations, and analysis files (MATLAB code) are in a common, intuitive file structure at https://searchworks.stanford.edu/view/zk881ps0522. The library may be used as course material or serve as a starter package for researchers early in their career or for established groups, to modify the analyses and re-apply them in
JRDB-Pose is a large-scale dataset and benchmark for multi-person pose estimation and tracking using videos captured from a social navigation robot. The dataset contains challenge scenes with crowded indoor and outdoor locations and a diverse range of scales and occlusion types. It provides human pose annotations with per-keypoint occlusion labels and tack IDs consistent across the scene. These annotations include 600,000 human body pose annotations and 600,000 head bounding box annotations.
PoseScript is a dataset that pairs a few thousand 3D human poses from AMASS with rich human-annotated descriptions of the body parts and their spatial relationships. This dataset is designed for the retrieval of relevant poses from large-scale datasets and synthetic pose generation, both based on a textual pose description.
This Kaggle repository is still under construction (as of October 2022).
The code to create the dataset is available here. The dataset used in the paper is available on github
S-TEST is a benchmark for measuring the specificity of the language of pre-trained language models.
ENTIGEN is a benchmark dataset to evaluate the change in image generations conditional on ethical interventions across three social axes -- gender, skin color, and culture. It contains 246 prompts based on an attribute set containing diverse professions, objects, and cultural scenarios.
arXivEdits an annotated corpus of 751 full papers from arXiv with gold sentence alignment across their multiple versions of revision, as well as fine-grained span-level edits and their underlying intentions for 1,000 sentence pairs. This dataset is designed for studying the human revision process in the scientific writing domain.
CodeSyntax is a large-scale dataset of programs annotated with the syntactic relationships in their corresponding abstract syntax trees. It contains 18,701 code samples annotated with 1,342,050 relation edges in 43 relation types for Python, and 13,711 code samples annotated with 864,411 relation edges in 39 relation types for Java. It is designed to evaluate the performance of language models on code syntax understanding.
Human single-cell atlas.
Parasitic infections have been recognized as one of the most significant causes of illnesses by WHO. Most infected persons shed cysts or eggs in their living environment, and unwittingly cause transmission of parasites to other individuals. Diagnosis of intestinal parasites is usually based on direct examination in the laboratory, of which capacity is obviously limited. Targeting to automate routine fecal examination for parasitic diseases, this challenge aims to gather experts in the field to develop robust automated methods to detect and classify eggs of parasitic worms in a variety of microscopic images. Participants will work with a large-scale dataset, containing 11 types of parasitic eggs from fecal smear samples. They are the main interest because of causing major diseases and illness in developing countries. We open to any techniques used for parasitic egg recognition, ranging from conventional approaches based on statistical models to deep learning techniques. Finally, the org
The dataset for this task is the TUT Urban Acoustic Scenes 2018 dataset, consisting of recordings from various acoustic scenes. The dataset was recorded in six large european cities, in different locations for each scene class. For each recording location there are 5-6 minutes of audio. The original recordings were split into segments with a length of 10 seconds that are provided in individual files. Available information about the recordings include the following: acoustic scene class, city, and recording location.