57 datasets found

The SUNRGB-D dataset features 5,285 training and 5,050 test samples from multiple RGB-D cameras. The dataset provides annotations for the first 37 NYUv2 semantic classes and for...

3D object detection datasets, ScanNet and SUNRGB-D

3D object detection is one of the most important tasks in autonomous driving and robotics. Our research focuses on tackling low efficiency issue of point-based methods on...

ULIP: Learning a unified representation of language, images, and point clouds for 3D understanding

The dataset used for training a Convolutional Neural Network to predict a grasp quality score over all grasp poses, given a depth image of an object.

The Argoverse 2 motion forecasting dataset contains 250,000 driving scenarios, each 11 seconds long. These scenarios cover 6 geographical regions and represent 763 total hours...

A Point-Cloud Deep Learning Framework for Prediction of Fluid Flow Fields on Irregular Geometries

Graph-ModelNet40, Graph-ModelNet10, Graph-ShapeNet Part are graph datasets constructed for graph classification task.

Recent Transformer-based 3D object detectors learn point cloud features either from point- or voxel-based representations.

A dataset for visual grounding on point clouds, focusing on relation modeling.

Self-driving dataset for scene flow estimation

The Waymo Open Dataset and the nuScenes Dataset are used to evaluate the performance of the AFDetV2 model.

The ShapeNet part dataset contains 16,881 shapes from 16 classes and 50 parts in total.

The ONCE dataset is a large-scale dataset for autonomous driving, containing 581 sequences composed of 20 labeled frames and 561 unlabeled frames.

A new 3D dataset, CC3D, a collection of 50k+ aligned pairs of meshes, a CAD model and its virtually 3D scanned counterpart with corresponding scanning artifacts.

MF: A proprietary mobile mapping platform dataset.

TUM-MLS-2016: An annotated mobile LiDAR dataset of the TUM City Campus for semantic point cloud interpretation in urban areas.

The dataset used in the paper is ScanNetV2, a real-world dataset for indoor scenes. It contains 1205 training scenes and 312 testing scenes, with instance-level object bounding...

SynthCity: A large scale synthetic point cloud dataset for pre-training deep learning models, enabling generalisation and expansion of their usage to real-world data.

The dataset used in the PointCLIP paper is ModelNet10, ModelNet40, and ScanObjectNN. ModelNet10 consists of 4,899 synthetic meshed CAD models with 10 indoor categories, 3,991...