Loading...
Loading...
3D models, rendered datasets, physics simulation, digital twins, synthetic data generation, game engine data
1,035 datasets
Tools and data for reconstructing 3D human avatars from unconstrained web-based video data as presented in CVPR 2025. This toolbox facilitates the creation of digital humans from 'in-the-wild' sources rather than controlled studio environments.
Curated by zshn25 and updated in November 2024, this repository aggregates 3D mesh model resources found across the web for research purposes. It serves as a centralized directory for 3D reconstruction and graphics data, indexing external repositories and web-based assets.
A synthetic dataset for converting natural language statements into Linear Temporal Logic (LTL) formulas, created by author cRick and published on Hugging Face in March 2023. It contains text data designed for formal methods and logic translation tasks. The dataset is categorized as having between 100,000 and 1 million entries.
Spectral-data and colorimetric resources for the Eclat Digital Ocean Light Simulator, developed by the colour-science organization. Updated in February 2023, this repository provides specialized spectral-data for marine light modeling and simulation.
Colour Substance provides color space definitions and configuration data for Allegorithmic Substance Painter and Designer, maintained by the colour-science organization. Updated in November 2024, the repository serves as a technical reference for color management within 3D texturing and material authoring workflows.
A synthetic text dataset created by 'myvision' for AI and machine learning applications. The dataset was last updated in June 2022 and is hosted on Hugging Face. Platform tags indicate it is a large-scale dataset, with size categories suggesting it contains between 100 million and 1 billion entries.
1 official repository for PointAR, a framework for efficient lighting estimation in mobile augmented reality presented at ECCV 2020. It provides resources for inferring spatially-varying lighting to improve the visual consistency of virtual objects in mobile environments.
3D point cloud samples across multiple object categories labeled for anomaly detection and localization tasks. This dataset establishes a benchmark for identifying spatial defects in 3D geometry as presented at NeurIPS 2023.
A dataset named Synthdog was authored by Jeongyeon and last updated on July 4, 2022. The dataset's modality includes image and text data, and it is associated with tags for the US region and libraries like mlcroissant, dask, and datasets. The specific number of rows, columns, and data size is unknown.
Known as titled 'Transparent' and was authored by HekmatTaherinejad. It was last updated on June 24, 2022. The dataset's modality is image, and it is tagged with the US region and a size category of 10K to 100K.
This repository aggregates multiple research papers, datasets, and code repositories focused on holistic 3D reconstruction within computer vision. The content is organized into categories such as scene understanding, object-level reconstruction, and geometric modeling.
Three resource types—research papers, datasets, and code repositories—are indexed to support 3D human body reconstruction research. This collection organizes academic literature alongside practical implementation links to facilitate the development of human modeling algorithms.
ModelNet10 contains 3D CAD models across 10 object categories, curated by SomTambe in May 2020. This version provides a compressed distribution of the standard Princeton ModelNet benchmark for 3D computer vision research.