Loading...
Loading...
3D models, rendered datasets, physics simulation, digital twins, synthetic data generation, game engine data
1,412 datasets
NASA's Kepler Mission, launched in March 2009, surveyed a portion of the Milky Way to discover Earth-sized planets in habitable zones. This package includes 3D models of the Kepler spacecraft, its photometer, and solar array, created by Ball Aerospace. The models are provided in PDF format.
Geoscience Australia's Marine Sediment Database (MARS) provides this dataset of sediment properties for over 15,000 seabed samples. Analytical data includes textural composition, particle size statistics, and calcium carbonate values, with location, depth, and survey information. The sample set spans the coast, shelf, slope, and deep ocean across the Australian marine region.
A synthetic dataset of paired indoor scene images for scene change detection research. Each pair shows the same scene before and after an object-level edit, rendered with NVIDIA Isaac Sim 5.1.0. The dataset includes RGB, depth, and segmentation modalities plus natural-language descriptions and corresponding action labels.
Qualcomm's synthetic dataset contains 25,087 mistake-intervention annotations for interactive cooking guidance. It includes video segments paired with instruction and feedback text and their timestamps. The data is derived from the CaptainCook4D, Ego4D, and Ego-Exo4D source datasets.
ClearDepth Transparent (FiftyOne) is a grouped dataset built from the synthetic transparent-object stereo data described in the ClearDepth paper. It contains indoor scenes with stereo RGB video, dense per-pixel labels, and optional merged 3D point-cloud reconstructions. The dataset was created by Voxel51.
Nine randomized controlled trials involving 944 children were included in this systematic review and meta-analysis. It evaluates the efficacy of virtual reality distraction for reducing needle-related procedural pain, anxiety, and fear in pediatric emergency departments, with data pooled from searches of PubMed, Embase, Cochrane Library, and Web of Science up to April 2024.
Montreal boroughs are represented by a digital 3D building model in CityGML LOD2 format with textures. The dataset includes building geometry and textures, and a related digital terrain model is available for a complete territorial representation. The data is provided in multiple formats including GML, DXF, and ZIP.
VideoASMR-Bench is a benchmark for evaluating the realism of AI-generated videos, specifically focusing on ASMR content. The dataset contains 100 test samples across two difficulty levels, created by author kolerk. It was last updated on May 5, 2026.
TeX-1500 is a paired dataset of 1,522 calibrated long-wave infrared hyperspectral images for temperature-emissivity-texture decomposition. The data was created by author jialelin2007 from DARPA pushbroom and WHU FTIR acquisitions. The repository was last updated on June 2, 2026, and currently contains a preview release with one example sample.
A study by Shaojin Ma, uploaded to figshare in 2026, employed computer vision to grade the age of test fabric used in washing machine performance standards. The dataset likely contains computational results for k-nearest neighbors, multilayer perceptron, linear discriminant analysis, and logistic regression classifiers trained on image features from fabric samples subjected to 1–100 accelerated washing cycles. The logistic regression model achieved accuracy scores ranging from 0.62 to 1.00 across five degradation stages.
LR classifier achieved accuracy of 0.75, 0.75, 0.62, 0.75, and 1.00 for five fabric degradation stages. This dataset contains image features extracted from base load fabric samples subjected to 1–100 accelerated washing cycles, used to investigate computer vision for aging assessment. It was created by Shaojin Ma and uploaded to figshare on April 10, 2026.
A dataset of 5.5 KB contains image-derived features from fabric samples used in washing machine performance testing. The data, authored by Shaojin Ma and updated in April 2026, includes color, texture, and area features extracted from wrinkle and weave information of base loads subjected to 1–100 accelerated washing cycles. It supports training classifiers like logistic regression for objective age grading.
86.4 KB of image-derived features from fabric samples undergoing 1–100 accelerated washing cycles. Shaojin Ma published this dataset on figshare in April 2026 to investigate computer vision techniques for grading fabric degradation. Features include color, texture, and area metrics extracted from wrinkle and plain weave structure information.
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. These models use the open-source ISO standard Virtual Reality Modelling Language (VRML) format for easy data sharing and interpretation.
Supplementary PDF files for a research article evaluating the business justification for implementing Digital Twins in legacy manufacturing systems. The files are associated with a dual-method study combining a systematic literature review and an industry survey of practitioners in the fast-moving consumer goods sector. The files were authored by Trev Bean and published on figshare in April 2026 under a CC-BY 4.0 license.
Imagery-derived point clouds classify elevation data into categories like ground and low noise. The data is organized into non-overlapping 1 km by 1 km tiles in a compressed format. The Government of Ontario maintains this dataset, with a last recorded update in March 2026.
UnrealMVS is a large-scale synthetic omnidirectional depth dataset rendered in Unreal Engine. It is designed for training and evaluating multi-view stereo networks, such as OmniMVS, on fisheye camera rigs. The dataset was created by flw-tu-dortmund and was last updated on 2026-05-22.
27 textured 3D scenes rendered for research in novel view synthesis and neural rendering. Each scene includes 401 frames captured along a predefined monocular camera trajectory. The dataset, created by Tengpaz, is a synthetic multi-scene rendering test for 3D-conditioned generative modeling.
SWOT Version C science data provides auxiliary information linking individual pixels to specific rivers and lakes. This point cloud data, covering tiles of approximately 64x64 km², includes height-constrained geolocation after reach- or lake-scale averaging. It is available in netCDF-4 format and supports the classification of water bodies from satellite observations.
SWOT Version C science data products provide a point cloud of water mask pixels over tiles approximately 64x64 km2. The data includes geolocated heights, backscatter, geophysical fields, and flags for each pixel. It is available in netCDF-4 format and represents a half swath from the SWOT instrument.