ShapeNet: An Information-Rich 3D Model Repository
arXiv:1512.03012
Abstract
We present ShapeNet: a richly-annotated, large-scale repository of shapes represented by 3D CAD models of objects. ShapeNet contains 3D models from a multitude of semantic categories and organizes them under the WordNet taxonomy. It is a collection of datasets providing many semantic annotations for each 3D model such as consistent rigid alignments, parts and bilateral symmetry planes, physical sizes, keywords, as well as other planned annotations. Annotations are made available through a public web-based interface to enable data visualization of object attributes, promote data-driven geometric analysis, and provide a large-scale quantitative benchmark for research in computer graphics and vision. At the time of this technical report, ShapeNet has indexed more than 3,000,000 models, 220,000 models out of which are classified into 3,135 categories (WordNet synsets). In this report we describe the ShapeNet effort as a whole, provide details for all currently available datasets, and summarize future plans.
Cited by in corpus (157)
- Image-based 3D Object Reconstruction: State-of-the-Art and Trends in the Deep Learning Era
- Point Transformer
- Learned Point Cloud Geometry Compression
- Comprehensive Review of Deep Learning-Based 3D Point Cloud Completion Processing and Analysis
- Deep Learning-based 3D Point Cloud Classification: A Systematic Survey and Outlook
- Unsupervised Point Cloud Representation Learning with Deep Neural Networks: A Survey
- Dense 3D Object Reconstruction from a Single Depth View
- Dark, Beyond Deep: A Paradigm Shift to Cognitive AI with Humanlike Common Sense
- Enable Deep Learning on Mobile Devices: Methods, Systems, and Applications
- 3D Sketching using Multi-View Deep Volumetric Prediction
- Adaptive O-CNN: A Patch-based Deep Representation of 3D Shapes
- A Rotation-Invariant Framework for Deep Point Cloud Analysis
- Synthetic Datasets for Autonomous Driving: A Survey
- Neural Dual Contouring
- ComplementMe: Weakly-Supervised Component Suggestions for 3D Modeling
- Synthesizing Diverse and Physically Stable Grasps with Arbitrary Hand Structures using Differentiable Force Closure Estimator
- Learning from Synthetic Shadows for Shadow Detection and Removal
- Deep Spatial Transformation for Pose-Guided Person Image Generation and Animation
- Joint Hand-object 3D Reconstruction from a Single Image with Cross-branch Feature Fusion
- Cross-Domain Complementary Learning Using Pose for Multi-Person Part Segmentation
- Hierarchical Attention Learning of Scene Flow in 3D Point Clouds
- PU-Net: Point Cloud Upsampling Network
- PhotoShape: Photorealistic Materials for Large-Scale Shape Collections
- Domain Adaptation on Point Clouds via Geometry-Aware Implicits
- GIFS: Neural Implicit Function for General Shape Representation
- Hypergraph Spectral Analysis and Processing in 3D Point Cloud
- 6D-ViT: Category-Level 6D Object Pose Estimation via Transformer-based Instance Representation Learning
- Gaussian Splatting: 3D Reconstruction and Novel View Synthesis, a Review
- CSDN: Cross-modal Shape-transfer Dual-refinement Network for Point Cloud Completion
- PRS-Net: Planar Reflective Symmetry Detection Net for 3D Models
- ROCA: Robust CAD Model Retrieval and Alignment from a Single Image
- Lossless Point Cloud Geometry and Attribute Compression Using a Learned Conditional Probability Model
- Advancements in Point Cloud-Based 3D Defect Detection and Classification for Industrial Systems: A Comprehensive Survey
- PVNAS: 3D Neural Architecture Search with Point-Voxel Convolution
- Robust Point Cloud Registration Framework Based on Deep Graph Matching(TPAMI Version)
- Are we done with object recognition? The iCub robot's perspective
- SceneDreamer: Unbounded 3D Scene Generation from 2D Image Collections
- A Comprehensive Review of Modern Object Segmentation Approaches
- SEG-MAT: 3D Shape Segmentation Using Medial Axis Transform
- Intrinsic Image Decomposition via Ordinal Shading
- DeltaConv: Anisotropic Operators for Geometric Deep Learning on Point Clouds
- P2C: Self-Supervised Point Cloud Completion from Single Partial Clouds
- Towards Real-World Category-level Articulation Pose Estimation
- Patch-Based Deep Autoencoder for Point Cloud Geometry Compression
- Pose from Shape: Deep Pose Estimation for Arbitrary 3D Objects
- Single-view 3D Mesh Reconstruction for Seen and Unseen Categories
- H-CNN: Spatial Hashing Based CNN for 3D Shape Analysis
- Spatial Transformer for 3D Point Clouds
- GRASPA 1.0: GRASPA is a Robot Arm graSping Performance benchmArk
- ClipGen: A Deep Generative Model for Clipart Vectorization and Synthesis
- FeatureNeRF: Learning Generalizable NeRFs by Distilling Foundation Models
- PSNet: Fast Data Structuring for Hierarchical Deep Learning on Point Cloud
- AUTO3D: Novel view synthesis through unsupervisely learned variational viewpoint and global 3D representation
- A Closer Look at Few-Shot 3D Point Cloud Classification
- CRA-PCN: Point Cloud Completion with Intra- and Inter-level Cross-Resolution Transformers
- SimJEB: Simulated Jet Engine Bracket Dataset
- Compact Model Representation for 3D Reconstruction
- DSGN: Deep Stereo Geometry Network for 3D Object Detection
- Optimal and Robust Category-level Perception: Object Pose and Shape Estimation from 2D and 3D Semantic Keypoints
- Robust Zero Level-Set Extraction from Unsigned Distance Fields Based on Double Covering
- AutoSweep: Recovering 3D Editable Objectsfrom a Single Photograph
- Neural-IMLS: Self-supervised Implicit Moving Least-Squares Network for Surface Reconstruction
- A Review of Emerging Research Directions in Abstract Visual Reasoning
- Virtual replicas of real places: Experimental investigations
- SoftPool++: An Encoder-Decoder Network for Point Cloud Completion
- DeepPoint3D: Learning Discriminative Local Descriptors using Deep Metric Learning on 3D Point Clouds
- CAD2Render: A Modular Toolkit for GPU-accelerated Photorealistic Synthetic Data Generation for the Manufacturing Industry
- A Comprehensive Survey of 3D Dense Captioning: Localizing and Describing Objects in 3D Scenes
- An Efficient Hypergraph Approach to Robust Point Cloud Resampling
- A Survey on Deep Generative 3D-aware Image Synthesis
- Deep Efficient Continuous Manifold Learning for Time Series Modeling
- Benchmarking Neural Radiance Fields for Autonomous Robots: An Overview
- Efficient Visual Recognition with Deep Neural Networks: A Survey on Recent Advances and New Directions
- Shallow2Deep: Indoor Scene Modeling by Single Image Understanding
- Scalable Surface Reconstruction with Delaunay-Graph Neural Networks
- Tinto: Multisensor Benchmark for 3D Hyperspectral Point Cloud Segmentation in the Geosciences
- Towards Analyzing Semantic Robustness of Deep Neural Networks
- SynH2R: Synthesizing Hand-Object Motions for Learning Human-to-Robot Handovers
- Learning to Group and Label Fine-Grained Shape Components
- NeeDrop: Self-supervised Shape Representation from Sparse Point Clouds using Needle Dropping
- Lidar Upsampling with Sliced Wasserstein Distance
- Learning Pose-invariant 3D Object Reconstruction from Single-view Images
- Learning Material-Aware Local Descriptors for 3D Shapes
- Enhancing Generalizable 6D Pose Tracking of an In-Hand Object with Tactile Sensing
- Learning Mesh Representations via Binary Space Partitioning Tree Networks
- Image Morphing with Perceptual Constraints and STN Alignment
- SnapshotNet: Self-supervised Feature Learning for Point Cloud Data Segmentation Using Minimal Labeled Data
- Regress Before Construct: Regress Autoencoder for Point Cloud Self-supervised Learning
- Dense Voxel 3D Reconstruction Using a Monocular Event Camera
- ELLIPSDF: Joint Object Pose and Shape Optimization with a Bi-level Ellipsoid and Signed Distance Function Description
- Beyond First Impressions: Integrating Joint Multi-modal Cues for Comprehensive 3D Representation
- Learning Modified Indicator Functions for Surface Reconstruction
- Benchmarking Convolutional Neural Network and Graph Neural Network based Surrogate Models on a Real-World Car External Aerodynamics Dataset
- Mitigating Shadows in Lidar Scan Matching using Spherical Voxels
- Parametric Surface Constrained Upsampler Network for Point Cloud
- Grasp Transfer based on Self-Aligning Implicit Representations of Local Surfaces
- CORSAIR: Convolutional Object Retrieval and Symmetry-AIded Registration
- PointSmile: Point Self-supervised Learning via Curriculum Mutual Information
- A Survey of Methods for Converting Unstructured Data to CSG Models
- Learning Online Visual Invariances for Novel Objects via Supervised and Self-Supervised Training
- RFTrans: Leveraging Refractive Flow of Transparent Objects for Surface Normal Estimation and Manipulation
- Shape Completion with Points in the Shadow
- Modular Interactive Video Object Segmentation: Interaction-to-Mask, Propagation and Difference-Aware Fusion
- Deep intrinsic decomposition trained on surreal scenes yet with realistic light effects
- Point Cloud Resampling Through Hypergraph Signal Processing
- KTNet: Knowledge Transfer for Unpaired 3D Shape Completion
- Generative AI-based Prompt Evolution Engineering Design Optimization With Vision-Language Model
- SuperPC: A Single Diffusion Model for Point Cloud Completion, Upsampling, Denoising, and Colorization
- Learning the Geodesic Embedding with Graph Neural Networks
- Ab Initio Particle-based Object Manipulation
- Incomplete Multimodal Industrial Anomaly Detection via Cross-Modal Distillation
- Semi-supervised Viewpoint Estimation with Geometry-aware Conditional Generation
- Generative Model with Coordinate Metric Learning for Object Recognition Based on 3D Models
- Combining Shape Completion and Grasp Prediction for Fast and Versatile Grasping with a Multi-Fingered Hand
- Online Adaptation for Implicit Object Tracking and Shape Reconstruction in the Wild
- Weakly Supervised Learning of Multi-Object 3D Scene Decompositions Using Deep Shape Priors
- ANISE: Assembly-based Neural Implicit Surface rEconstruction
- Composite Convolution: a Flexible Operator for Deep Learning on 3D Point Clouds
- ExpPoint-MAE: Better interpretability and performance for self-supervised point cloud transformers
- Joint Data and Feature Augmentation for Self-Supervised Representation Learning on Point Clouds
- SynBench: A Synthetic Benchmark for Non-rigid 3D Point Cloud Registration
- Object Segmentation of Cluttered Airborne LiDAR Point Clouds
- TetraSphere: A Neural Descriptor for O(3)-Invariant Point Cloud Analysis
- Hierarchical Point Cloud Encoding and Decoding with Lightweight Self-Attention based Model
- GenPara: Enhancing the 3D Design Editing Process by Inferring Users' Regions of Interest with Text-Conditional Shape Parameters
- SceneEDNet: A Deep Learning Approach for Scene Flow Estimation
- HgPCN: A Heterogeneous Architecture for E2E Embedded Point Cloud Inference
- Latent-Space Laplacian Pyramids for Adversarial Representation Learning with 3D Point Clouds
- Shape Completion with Prediction of Uncertain Regions
- Robust-DefReg: A Robust Deformable Point Cloud Registration Method based on Graph Convolutional Neural Networks
- Denoising-While-Completing Network (DWCNet): Robust Point Cloud Completion Under Corruption
- LiveNVS: Neural View Synthesis on Live RGB-D Streams
- CAD-NeRF: Learning NeRFs from Uncalibrated Few-view Images by CAD Model Retrieval
- Self-supervised Learning of Rotation-invariant 3D Point Set Features using Transformer and its Self-distillation
- Learning Self-Prior for Mesh Inpainting Using Self-Supervised Graph Convolutional Networks
- IPC-Net: 3D point-cloud segmentation using deep inter-point convolutional layers
- Bent & Broken Bicycles: Leveraging synthetic data for damaged object re-identification
- Learning Depth With Very Sparse Supervision
- Local region-learning modules for point cloud classification
- Gap Completion in Point Cloud Scene occluded by Vehicles using SGC-Net
- 3D Magic Mirror: Clothing Reconstruction from a Single Image via a Causal Perspective
- TIDE: Temporally Incremental Disparity Estimation via Pattern Flow in Structured Light System
- DeepDiffusion: Unsupervised Learning of Retrieval-adapted Representations via Diffusion-based Ranking on Latent Feature Manifold
- 360° Stereo Image Composition with Depth Adaption
- Representation Learning of Point Cloud Upsampling in Global and Local Inputs
- Winding Clearness for Differentiable Point Cloud Optimization
- Human Inspired Progressive Alignment and Comparative Learning for Grounded Word Acquisition
- Understanding Pixel-level 2D Image Semantics with 3D Keypoint Knowledge Engine
- Approaching human 3D shape perception with neurally mappable models
- Scan2Part: Fine-grained and Hierarchical Part-level Understanding of Real-World 3D Scans
- RECALL: Rehearsal-free Continual Learning for Object Classification
- CageNet: A Meta-Framework for Learning on Wild Meshes
- Shape2.5D: A Dataset of Texture-less Surfaces for Depth and Normals Estimation
- Augmented Environment Representations with Complete Object Models
- Component Selection for Craft Assembly Tasks
- Fuse Before Reconstruct: A Novel Reconstruction Paradigm for Multimodal Industrial Anomaly Detection
- Diff-3DCap: Shape Captioning with Diffusion Models