Loading...
Loading...
3D models, rendered datasets, physics simulation, digital twins, synthetic data generation, game engine data
1,408 datasets
75 steady 3D RANS solutions for a 14-blade axial fan rotor passage across its operating map. The dataset contains flow fields for parametric blade geometries derived from six real-world primary rotors via Parablade, created by JeoaFesketto and last updated in July 2026. Computations were performed using the FINE/Turbo v18 (Euranus) solver.
Magnetic Resonance Imaging (MRI) data from a male Wistar rat was used to create a realistic 3D mechanical model. The data consisted of 240 cross-sectional slices, each 0.22mm thick, with a resolution of 320x320 voxels at 0.3mm. The model was processed into IGES and .stl files by Bernd M. Pohl for physical replica creation.
Hypershadow is a benchmark dataset designed to answer one specific question: given a 3D point cloud, can you tell if it is an ordinary 3D object or the 3D projection of an object from a higher spatial dimension. The dataset was created by AkshaySasi and was last updated on Hugging Face in July 2026. Label 1 clouds are projections of objects living in R^4, R^5, or R^6, including hyperspheres, tesseracts, and Clifford tori.
Research Inventy conducted a study examining production factors for mozzarella cheese from cow's milk. The work involved three experiments per milk kind, with three patterns analyzed per experiment, tracking processes from cheese drying to market preparation. The objective was to apply production technology and supply Kosovo's trade market with locally produced mozzarella cheese.
Three 3D model files of sphere dice in the .stl format, created by Mitchell J. Feldmann at the University of California, Davis. The models were modified through scaling from the Sphere Dice dataset by anvil777, originally sourced from Thingiverse.
75 women selected among 33 three-dimensional models to assess size recall accuracy and preferences for erect penis size in different relationship contexts. The study, authored by Nicole Prause, found women recalled model size accurately but made more errors with length than circumference. Preferences indicated a slightly larger size for one-time versus long-term partners.
Model results from five Bern3D Earth system simulations published by Frerk Pöppelmeier at the University of Bern in 2021. The datasets compare pre-industrial conditions with various Last Glacial Maximum scenarios, including changes to ocean circulation and wind patterns.
Model-based predictions for four deer species across Victoria, based on surveys at 317 sites between 2021 and 2023. The raster data provides three confidence levels for occupancy per square kilometre: smallest, average, and largest estimated ranges. This dataset was produced by the Arthur Rylah Institute and published by the Department of Energy, Environment and Climate Action.
2.3 GB of data files implementing a framework for generating high-fidelity 3D tree mesh models from laser point clouds. The dataset, authored by Hai Xu and associated with a 2026 journal article, includes code and outputs for a method that integrates data-driven and morphology-knowledge-driven techniques to address challenges like leaf occlusion and data gaps. It was last updated on 2026-05-18.
A research paper and associated materials present findings from a user study on art-inspired visualizations in Virtual Reality. The study derived design considerations from art forms, applied them to common charts, and gathered feedback on usability, engagement, and clarity. The work was authored by Arun Kumar Sekar and last updated in May 2026.
OmniScene is a photorealistic synthetic dataset rendered in NVIDIA Isaac Sim. It was created by henryzhou998 to train the X-Lens model and was last updated on July 11, 2026. Each scene is captured by a 6-camera rig consisting of 4 fisheye and 2 pinhole cameras, providing metric ground-truth depth.
69 first-year Computer Science students were randomly assigned to three groups to complete drawing tasks using a GUI, a physical robot, or a Digital Twin. The dataset includes anonymized participant data, group assignments based on prior performance bands, and task evaluation results. The study was conducted by Michael Heeney and published on figshare in 2026.
10 participants with mild to borderline intellectual disability (MBID) in a forensic psychiatric setting were part of a pilot study investigating a Virtual Reality Aggression Prevention Training (VRAPT-ID). The dataset likely contains physiological measures like pulse rate and electrodermal activity, alongside behavioral assessments using the Social Dysfunction and Aggression Scale (SDAS). The study was authored by Patricia van Reekum and last updated on 2026-05-29.
A 9.8 MB collection of research materials on the Degree-Corrected Mixed Membership (DCMM) model for network data. The dataset, authored by Sohom Bhattacharya and last updated in May 2026, includes PDF and ZIP files containing theoretical derivations, a ranking scheme for vertices, and numerical experiments on real and synthetic data. It provides methods for asymptotic distributions, confidence intervals, and a multiplier bootstrap for ranking inference of individual membership profiles.
A five-day voyage from New Zealand to Tasmania in June 2009 collected biological and acoustic data from the Tasman Sea. Sampling involved a MIDOC net system deployed to 1000 meters, a 38 kHz echosounder logged for the entire transit, and an Acoustic-Optical System (AOS) deployed to 1000 meters. The data was collected by CSIRO Marine and Atmospheric Research during a 24-hour charter of the FV Rehua.
5.2 GB of texture images generated by a text-to-image AI model to evaluate tactile roughness perception. Yui Momiyama created this dataset to investigate whether visual features can quantify tactile roughness and how linguistic prompts influence the generated textures. The dataset, last updated in May 2026, includes images across wood, metal, stone, fiber, plastic, and ceramic categories.
Geoscience Australia produced free Web-viewable 3D models integrating coastal spatial data. The models combine digital elevation models, multibeam bathymetry, sediment samples, benthic habitats, and satellite imagery for the Keppel Bay and Fitzroy River area in Queensland, Australia. They use the open-source ISO standard Virtual Reality Modelling Language (VRML) file format for easy web distribution.
PointNetGPD-based scores for 47 single-view point clouds of common objects. The dataset includes positive grasping pose samples and experimental results from the paper 'An efficient pose classification method for robotic grasping'. Author wenlong ji released the data on paperswithcode.
Synthetic collateral mix data shows how value, liquidity, and seasonality differ across core pawn inventory categories and subcategories. This build contains 48 rows under the seasonal back to school scenario. The dataset was created by gene fur and published on figshare on 2026-04-12.
A first-iteration 3D geological model integrates interpretation of approximately 1,580 km of deep seismic reflection data, 40,000 line kilometres of AusAEM airborne electromagnetic data, and stratigraphic review of more than 800 legacy boreholes. Developed by Geoscience Australia as part of the Resourcing Australia’s Prosperity initiative, this basin-scale model provides a representation of stratigraphic architecture and basin geometry. The associated data package includes 22 regional stratigraphic surfaces, 29 isochore grids, 16 median Vshale maps, and 20 confidence maps.