Thingi10K: A Dataset of 10, 000 3D-Printing Models

By Qingnan Zhou, Alec Jacobson. arXiv.org, 2016

537 citations recorded by the source corpus.

Empirically validating new 3D-printing related algorithms and implementations requires testing data representative of inputs encountered \emph{in the wild}. An ideal benchmarking dataset should not only draw from the same distribution of shapes people print in terms of class (e.g., toys, mechanisms, jewelry), representation type (e.g., triangle soup meshes) and complexity (e.g., number of facets), but should also capture problems and artifacts endemic to 3D printing models (e.g., self-intersections, non-manifoldness). We observe that the contextual and geometric characteristics of 3D printing models differ significantly from those used for computer graphics applications, not to mention standard models (e.g., Stanford bunny, Armadillo, Fertility). We present a new dataset of 10,000 models collected from an online 3D printing model-sharing database. Via analysis of both geometric (e.g., triangle aspect ratios, manifoldness) and contextual (e.g., licenses, tags, classes) characteristics, we demonstrate that this dataset represents a more concise summary of real-world models used for 3D printing compared to existing datasets. To facilitate future research endeavors, we also present an online query interface to select subsets of the dataset according to project-specific characteristics. The complete dataset and per-model statistical data are freely available to the public.

Subfields: Automotive Engineering, Computational Mechanics, Computer Graphics and Computer-Aided Design

View Paper: https://www.semanticscholar.org/paper/f558cda8c3065dba772ce6b21741bb283ca0ea4e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Automotive Engineering

In Computational Mechanics

In Computer Graphics and Computer-Aided Design

Citations in the corpus

ABO: Dataset and Benchmarks for Real-World 3D Object Understanding (2021, background): “One way around the challenging problem of getting 3D annotations for real images is to focus only on synthetic, computer-aided design (CAD) models [5, 27, 67].”
Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapes (2021, methodology): “ShapeNet150 and Thingi32 are evaluated using Chamfer-L 1 distance (multiplied by 10 3 ) and intersection over union over the uniformly sampled points (gIoU).”
Shape As Points: A Differentiable Poisson Solver (2021, methodology): “We consider three different types of 3D inputs: point clouds sampled from synthetic meshes [74] with Gaussian noise, real-world scans [68], and high-resolution raw scans of humans with comparably little noise [6].”
ABC: A Big CAD Model Dataset for Geometric Deep Learning (2018, background): “#Models C AD F il e s C u r v e s P a t c h e s S e m a n ti c s C a t e go r i e s ABC 1,000,000+ – – ShapeNet* [21] 3,000,000+ – – – ShapeNetCore 51,300 – – – ShapeNetSem 12,000 – – – ModelNet [54] 151,128 – – – Thingi10K [58] 10,000 – – – – PrincetonSB[10] 6670 – – – – NIST [3] ≤ 30 – –”
Objaverse-XL: A Universe of 10M+ 3D Objects (2023, background): “More recently, IKEA [20] 219 GSO [21] 1K EGAD [22] 2K OmniObject3D [23] 6K PhotoShape [24] 5K ABO [25] 8K Thingi10K [26] 10K 3d-Future [27] 10K ShapeNet [13] 51K Objaverse 1.0 [14] 800K Objaverse-XL 10.2M Segment Anything (SAM) [28] introduced a dataset of one billion object masks used to train a…”
Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation (2025, unknown): “From 3Dscanrep [18, 44, 93] to ShapeNet [8], the growth of 3D datasets has been gradual [85, 123, 25, 77, 103, 21, 17].”
Text2Mesh: Text-Driven Neural Stylization for Meshes (2021, background): “We consider a variety of sources including: COSEG [54], Thingi10K [71], Shapenet [5], Turbo Squid [57], and ModelNet [59].”