Loading...
Loading...
3D models, rendered datasets, physics simulation, digital twins, synthetic data generation, game engine data
1,020 datasets
A synthetic dataset designed to mimic comprehensive medical check-up results, including medical history, physical exams, lab results, vaccinations, and occupational risk factors. The dataset was created by author 'udeezz' and was last updated on April 27, 2026. Its specific scale in terms of rows and columns is not detailed in the provided metadata.
VBVR-Dataset provides 1,000,000 video clips organized into 100 curated reasoning task generators, released by Video-Reason in February 2026. Each record pairs a video segment with start/end frame indices and a textual reasoning prompt to facilitate spatiotemporal understanding.
LIBERO GT Depth Aligned Hide Sites Fast contains ground-truth depth sidecar files for LIBERO robot demonstrations. Each episode is stored as a compressed .npz file, with depth aligned to original frames by re-rendering from simulator states. The dataset was authored by SeonghuJeon and last updated on 2026-04-22.
Vista4D is an evaluation dataset for video reshooting with 4D point clouds, associated with a CVPR 2026 Highlight paper. The dataset was created by a consortium including Eyeline Labs, Netflix, Columbia University, UCLA, Stony Brook University, and the University of Oxford. It was last updated on HuggingFace on April 24, 2026.
122,199 high-quality, synthetic Question and Answer pairs are designed to instruction-tune Large Language Models into expert coding and architectural assistants for Unreal Engine 5.7. The dataset was created by TunstallTensor and last updated on Hugging Face in April 2026. It specifically addresses the frequent API deprecations and paradigm shifts introduced in the game engine.
Southern California coastal waters contain data from 426 fisheries-independent benthic trawls conducted by the Southern California Coastal Water Research Project (SCCWRP). The dataset originally included 288 invertebrate species, but analysis focused on 41 species present in at least 5% of 401 trawls, collected at depths from 2 to 215 meters during June-August. Site clusters were calculated using the Bray-Curtis dissimilarity coefficient.
G-LiHT airborne LiDAR point cloud data provides high-density individual return information for terrestrial ecosystems. The data includes 3D coordinates, ground classifications, height, and reflectance, with a density exceeding 10 points per square meter. NASA produced this data, last updated in March 2026.
MODIS/Terra+Aqua Burned Area data provides monthly, global gridded maps of fire-affected land at a 500-meter spatial resolution. The product is generated by NASA's LPCLOUD using a combined algorithm from MODIS Surface Reflectance and active fire observations. It identifies the specific burn date for each pixel within a calendar year.
Proactive Agent Dataset is a collection for training and evaluating proactive IDE assistants. It contains 10,489 synthetic samples from a gym simulation and 4,346 real IDE sessions split into training, validation, and test sets. The dataset was created by author jiarx and last updated on April 20, 2026.
SynCom (Synthetic Composition) is a synthetic dataset for realistic 3D object-scene composition. The dataset, authored by yGaoJiany, provides controlled object placement, known camera parameters, and rendered ground truth for composed scenes. It was last updated on April 29, 2026.
1,138 synthetic furniture images with detailed annotations, created by MohamedSameh77i. The dataset was last updated on May 16, 2026. It is hosted on the Hugging Face platform.
Launched in April 2006, CloudSat monitors the Earth's atmosphere with a radar system developed jointly with the Canadian Space Agency. The dataset likely contains 3D models and radar data for observing cloud moisture, predicting rain-producing clouds, and monitoring snowfall.
Synthetic data simulating experimental results from 60 participants performing target identification tasks under varying AI autonomy conditions. The dataset supports the study 'Sociotechnical Challenges of Defence AI: Automation Bias, Human Control and Accountability' and was generated by Shubham Soni. It was last updated on April 14, 2026.
NOAA_NCEI Accession 0176511 contains conductivity, temperature, depth (CTD) and plankton relative abundance data from the R/V Bellows cruise BE-1403. The dataset covers 27 CTD sample sites and 14 plankton net tows collected off the Florida Panhandle in the DeSoto Canyon from September 14-16, 2013. Plankton counts are categorized into 12 taxonomic groups, with abundances recorded as both raw counts and percentages.
Nghia Vu's research dataset for affordance learning in 3D scenes, presented at CVPR 2026. The dataset likely contains visual and geometric data to identify interactable regions in complex environments. It was uploaded to Hugging Face by the author aiozai and last updated on April 27, 2026.
Photos document 3D-printed prototypes for armports and viewports. Casting mold tests are also featured, along with part assembly layouts. James Lee contributed the dataset, which includes .STL 3D model files of the armport and viewport, and it was last updated on May 9, 2026.
STEP 3D model files for the 2.5 iteration of a slider were created by James Lee. The files feature refined designs intended for in-house manufacturing methods and utilize common neodymium magnets. The collection also includes a 3D-printable jig for drilling holes into magnet holders made from HDPE flat stock.
A 1500-year chronology of beach-ridge morphology, sediment texture, and geochemistry from Keppel Bay, Queensland. The dataset was built using optically stimulated luminescence (OSL) dating and is provided by Geoscience Australia Data, last updated in March 2026. Preliminary results suggest changes in shoreline accumulation rates, sediment sources, and minor relative sea-level falls.
8,675 images across 64 texture classes, preprocessed to 128x128 pixels. This dataset is a curated combination of four established texture datasets (KTH-TIPS, KTH-TIPS 2, Kylberg, and UIUC) collected in laboratory conditions. It was compiled and released by Ihsan Ullah for the Meta-Album project in March 2022.
A 42.4 KB Excel file from figshare evaluates uplink performance in Terahertz networks. Author Tahniyat Aslam compares 3D and 2D antenna models under realistic blockage conditions from humans and walls. The dataset was last updated on April 9,我们发现了一个问题，日期格式不符合要求，已修正。