Loading...
Loading...
3D models, rendered datasets, physics simulation, digital twins, synthetic data generation, game engine data
1,029 datasets
A dataset from the Natural Scenes Dataset (NSD) project, likely containing pre-selected voxel data for training models. It is hosted on the Kaggle platform. The specific content, scale, and collection methodology require verification after download.
NSD S6 val train NC selected voxels 70 is a dataset hosted on Kaggle. The title suggests it contains selected voxel data, likely from functional magnetic resonance imaging (fMRI), for training models related to the Natural Scenes Dataset. The specific content, scale, and origin require verification after download.
A synthetic dataset likely modeling customer interactions with add-on items in a food delivery cart, sourced from Kaggle. The dataset's specific size, creator, and temporal coverage are not provided in the metadata. Its content and structure must be verified after download.
Virtualized GPU Rendering Scheduling Data focuses on real-time resource allocation for graphics processing units. The dataset, sourced from Kaggle, likely contains metrics related to scheduling decisions and performance under virtualization. Its specific scale, authorship, and update history are not detailed in the provided metadata.
A dataset of extracted neural textures, likely for use in computer graphics and machine learning. The dataset is published on Kaggle, but specific details about its size, creation date, and author are not provided. Its content and structure must be verified after download.
3,000 synthetically generated image-text pairs created using the Tongyi-MAI/Z-Image-Turbo (v1.0) model. Each entry consists of a high-resolution image and a corresponding text prompt file. The dataset was authored by lrzjason and last updated on Hugging Face in November 2025.
Historic Area Remediation Site (HARS) offshore New York and New Jersey contains sea floor sediment texture data from 1996 and 1998 surveys. The dataset includes maps of sea floor topography, shaded relief, and backscatter intensity at a 3-meter per pixel resolution, showing changes from dredged material placement. This study was conducted cooperatively by the U.S. Geological Survey and the U.S. Army Corps of Engineers.
Snow depth maps with a spatial resolution of 0.5 meters are derived from images captured by a Vexcel Ultracam camera mounted on a piloted airplane. The maps are calculated by subtracting a summer digital terrain model from a winter digital surface model, with extensive masking applied to increase reliability. The dataset is published by ENVIDAT and hosted on NASA EarthData, with accuracy assessments showing a root mean square error of 0.25 m for 2017 and 0.15 m for subsequent years.
MODIS/Terra+Aqua Direct Broadcast Burned Area Monthly L3 Global 500m SIN Grid V006 provides monthly, global burned-area detection at a 500-meter spatial resolution. The product was decommissioned in July 2023, with users directed to the updated Version 6.1. It was produced by LPCLOUD using combined data from NASA's Terra and Aqua satellites.
Mesozooplankton grazing rates were measured during the 1992 Joint Global Ocean Flux Study Equatorial Pacific Process Study. Data were collected from 43 station occupations across four research cruises along 140°W longitude, from 17°S to 12°N. The dataset was produced by the SCIOPS organization and captures measurements from February to October 1992.
NOAA_NCEI Accession 0176344 reports plankton species abundance from 14 sites and quantitative phytoplankton profiles from 5 sites in the Gulf of Mexico's DeSoto Canyon. Data was collected aboard the R/V Bellows cruise BE 1317 from April 6 to 8, 2013, using net tows and Niskin bottle sampling with subsequent scanning electron microscope analysis. The dataset includes species counts, percentages, cell abundances per liter, and associated micrographs.
ScreenSpot-V2 is a FiftyOne dataset containing 1,272 samples, as stated in its dataset card. The dataset was created by Voxel51 and was last updated on Hugging Face on June 25, 2025. Its specific content must be inspected via the FiftyOne app or the full description on its Hugging Face page.
800 receipt image samples are available for analysis, curated by Voxel51 and last updated in October 2025. The dataset is intended for tasks in document analysis and optical character recognition.
90,289 raw and color-corrected GoPro images documenting coral reefs in Indonesia for 3D photogrammetry. Created by Wildflow and updated in April 2025, the collection provides high-resolution visual data for marine environment reconstruction.
TexVerse is a large-scale 3D dataset featuring high-resolution textures. It contains 858,669 unique 3D models curated from Sketchfab, including 158,518 models with specific characteristics. The dataset was created by researchers from Shanghai Innovation Institute, Jilin University, Fudan University, and Tongji University, and was last updated on September 3,我们发现了一个问题。
The M2I dataset is a synthetic dataset introduced in the paper 'MIC-BEV: Multi-Infrastructure Camera Bird's-Eye-View Transformer with Relation-Aware Fusion for 3D Object Detection'. It is designed to support training and evaluation of models for infrastructure-based multi-camera 3D object detection. The dataset was uploaded by author 'handsomeYun' and last updated on December 13, 2025.
Research abstracts analyze the poly-metamorphic terrain and structural evolution of the East Antarctic mobile belt in the Prydz Bay area. The data originates from the Australian Antarctic Data Centre and was last updated in March 1994. It compiles findings on deformation phases, metamorphic conditions, and kinematic interpretations from geological studies.
43.2 km² of a high-alpine catchment in the Swiss Alps was simulated for the 2014-2015 water year using the Alpine3D model. The dataset, from ENVIDAT, combines multi-scale field data including snow lysimeter measurements, distributed snow depths, and streamflow to validate model performance. It was published in 2017.
Zooplankton samples were collected via oblique bongo net tows to a depth of 20 meters in the coastal waters of the Gulf of Alaska. The survey occurred aboard the F/V Great Pacific from October 19 to November 8, 2004, using nets with 335 and 505 micron mesh. Samples were preserved in formalin onboard and analyzed in a laboratory.
Zooplankton counts were collected via Tucker trawl surveys in the coastal waters of the Gulf of Alaska from 2001 to 2003. Surveys were conducted aboard the F/V Great Pacific in 2001-2002 and NOAA Ship Miller Freeman in 2003, sampling transects from nearshore to beyond the 200-meter shelf break. Plankton samples were preserved in formalin onboard and analyzed in a laboratory.