Dataset produced for the SAPIEN simulation environment. From the website: "PartNet-Mobility dataset is a collection of 2K articulated objects with motion annotations and rendernig material. The dataset powers research for generalizable computer vision and manipulation. The dataset is a continuation of ShapeNet and PartNet. "

yuchen0187/partnet-mobility-test dataset hosted on Hugging Face and contributed by the HF Datasets community

From PARIS: Part-level Reconstruction and Motion Analysis for Articulated Objects: 5.1. Dataset Synthetic dataset. The synthetic 3D models we use for evaluation are from the PartNet-Mobility dataset [49, 27, 4], a large-scale dataset for articulated objects across 46 categories. We select instances across 10 categories to conduct our experiments. For each articulation state, we randomly sample 64-100 views covering the upper hemisphere of the object to simulate capturing in the real world. Then we render RGB images and acquire camera parameters and object masks using Blender [6] to create our training data. Real-world dataset. The real data we use for experiments is from the MultiScan dataset [25], scanning real-world indoor scenes with articulated objects in multiple states. We use the reconstructed mesh of an object in two states as ground truth for evaluation, and the real RGB frames as training data.

From the project GitHub: We release both synthetic and real data shown in the paper here. Once downloaded, folders data and load should be put directly under the project directory. If you find it slow to download the data from our server, please try this OneDrive link instead.

This repository contains the synthetic data used in the paper Generalizable Articulated Object Reconstruction from Casually Captured RGBD Videos

  Term of Use

Our dataset is derived from the PartNet-Mobility dataset. Users are required to agree on the terms of use of the PartNet-Mobility dataset before using our dataset. Researchers shall use our dataset only for non-commercial research and educational purposes.

  File Structure

Inside the sim_data folder, there are several… See the full description on the dataset page: https://huggingface.co/datasets/3dlg-hcvc/video2articulation.

ArtImage is a synthetic dataset of articulated object models of 5 categories from PartNet-Mobility for articulated object tasks in category level.

PartNet is a consistent, large-scale dataset of 3D objects annotated with fine-grained, instance-level, and hierarchical 3D part information. The dataset consists of 573,585 part instances over 26,671 3D models covering 24 object categories. This dataset enables and serves as a catalyst for many tasks such as shape analysis, dynamic 3D scene modeling and simulation, affordance analysis, and others.

This repo contains the data for S2O: Static to Openable Enhancement for Articulated 3D Objects. See the code on GitHub and the paper for details. Please cite S2O [1] if you use ACD. We provide the mesh, point cloud, and metadata for the two datasets used in S2O.

PM-Openable - This is a subset of 648 openable objects from full PartNet-Mobility [2]. We use a train/val/test split of 460/95/93 objects.
Articulated Container Dataset (ACD) [1] - We take openable container objects from HSSD [3]… See the full description on the dataset page: https://huggingface.co/datasets/3dlg-hcvc/s2o.

The link includes both our OPDSynth and OPDReal dataset. For OPDSynth, we select objects with openable parts from an existing dataset of articulated 3D models PartNet-Mobility. For OPDReal, we reconstruct 3D polygonal meshes for articulated objects in real indoor environments and annotate their parts and articulation information.