Loading...
Loading...
3D models, rendered datasets, physics simulation, digital twins, synthetic data generation, game engine data
1,034 datasets
Aggregating synthetically generated data used to train the Architext models for language-driven generative architecture design. The data was created using a parametric design script in Grasshopper 3D. The dataset is categorized for text generation tasks and was last updated in May 2023.
Manitoba soil texture data collected for the Soil Moisture Active Passive Validation Experiment 2016 Manitoba (SMAPVEX16 Manitoba) campaign. The data set was last updated on July 31, 2016. It is hosted by NSIDC_CPRD on the NASA Earthdata platform.
SMAPVEX12 soil texture data was collected using coring devices at multiple sites. The data set supports validation of satellite-based soil moisture measurements. It was produced by NSIDC_CPRD and last updated in July 2012.
Analcime, an unusual mineral, forms a pure horizon traced along the Mt. Fleming ridge, suggesting a distal Permian volcanic ash event. The dataset documents an investigation and extensive sampling of the lower Wellar Coal Measure sequence by organization SCIOPS. Data was last updated in January 1977.
CLASIC07 Soil Texture Data V001 contains soil texture information extracted for the Cloud and Land Surface Interaction Campaign 2007. Data were generated from the CONUS-Soil database for the conterminous United States, representing conditions during the CLASIC07 campaign timeline. The dataset was produced by the NSIDC_CPRD organization and released in July 2007.
34,814 polygons and 19,011 vertices define this detailed 3D model of the Saturn V rocket. The model was published by the National Aeronautics and Space Administration and last updated in April 2025. It is provided in the IMAGE/X-3DS file format.
A 3D model of a NASA Space Shuttle Orbiter contains 5054 polygons and 3441 vertices. The model is provided in the IMAGE/X-3DS file format and was last updated in April 2025.
216,930 Jeopardy! game show questions and answers spanning multiple decades of television history. Each entry includes metadata such as the category, dollar value, round type, show number, and original air date.
3D models of the radius, ulna, forearm rotation axis, and rotational centers at the DRUJ and PRUJ joints. The models were obtained from 3D- and 4D-CT acquisitions in healthy volunteers. The dataset was authored by Joris Oonk and last updated on October 15, 2025.
211 human-authored 3D scenes and over 18,000 models of real-world objects comprise the Habitat Synthetic Scenes Dataset (HSSD). This dataset is designed to more closely mirror real interiors than prior synthetic scene collections. The dataset was created by 'hssd' and was last updated on the Hugging Face platform in June 2023.
8,320 data samples of uniformly distributed points and their signed distance function (SDF) values. The subset contains three shape classes: airplanes (2156 samples), chairs (4189 samples), and sofas (1975 samples). It was created by AlexWolski and last updated on November 30, 2022.
Moroccan Darija synthetic dataset for transliteration tasks. The dataset, created by Haitam03, provides Latin script input paired with Arabic script output and normalized forms. It was last updated on Hugging Face on October 29, 2025.
New York City land cover data provides an 8-class classification derived from 2017 LiDAR and 2016 orthoimagery, created for an urban tree canopy assessment. The dataset uses a 'top-down' mapping perspective where overhanging canopy is assigned to the Tree Canopy class, with vegetation below 8 feet classified as Grass/Shrub.
A synthetic dataset of question-answer pairs extracted from the A-Roucher/huggingface_doc repository. The dataset was created by author m-ric and last updated on July 3, 2024. It is intended for evaluating the performance of Retrieval-Augmented Generation (RAG) systems.
A stereo video dataset of in vivo endoscopy from the Hamlyn Center Laparoscopic at Imperial College London. It provides rectified stereo images, calibration data, and ground truth depth maps generated using the Libelas stereo matching software. The dataset was uploaded by Recasens and last updated on 2024-09-25.
A research paper archived as a PDF, authored by Kathryn Mesh in 2021. The publication investigates how the height of pointing gestures marks target distance among speakers of the San Juan Quiahije Chatino language. The paper was published in the journal Lingua, volume 259.
AdamCodd processed a dataset focused on the code from Unreal Engine 5. The dataset was last updated on June 17, 2024. Its specific size, format, and structure are not detailed.
RoboCasa provides a collection of 3D assets adapted for the ManiSkill and SAPIEN simulation platforms. The dataset includes corrected material files for objects like walls, outlets, toasters, and fridges to address visual rendering issues. It was uploaded by haosulab and last updated on October 24, 2024.
900 simulated tax submissions contain 20 different 1988 IRS form faces, averaging 6.22 forms per submission. The National Institute of Standards and Technology created this binary image database for testing form recognition systems, using computer-derived images that mimic real hand-printed forms but contain no actual taxpayer data.
11,141 rigged 3D models across categories such as animals, humans, and furniture. Data includes 3D meshes, hierarchical skeletons, and per-vertex skinning weights for training automated rigging systems.