ShapeNet: An Information-Rich 3D Model Repository

2023/11/09 by Angel X. Chang, Thomas Funkhouser, Leonidas Guibas +10 · 1 citation
Engineering · Computer Science · #3D Shape Modeling and Analysis #Image Processing and 3D Reconstruction #Manufacturing Process and Optimization

paper · doi:10.5281/zenodo.10089705

openalex publication_date 2023/11/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/16

Abstract

Authors listed alphabetically We present ShapeNet: a richly-annotated, large-scale repository of shapes represented by 3D CAD models of ob-jects. ShapeNet contains 3D models from a multitude of semantic categories and organizes them under the Word-Net taxonomy. It is a collection of datasets providing many semantic annotations for each 3D model such as consis-tent rigid alignments, parts and bilateral symmetry planes,

Cited by

• SoftFlow: Probabilistic Framework for Normalizing Flow on Manifolds
• Probabilistic Category-Level Pose Estimation via Segmentation and Predicted-Shape Priors
• StablePose: Learning 6D Object Poses from Geometrically Stable Patches
• RGB-D Local Implicit Function for Depth Completion of Transparent Objects
• Self-Supervised Multi-View Learning via Auto-Encoding 3D Transformations
• Visualizing the Invisible: Occluded Vehicle Segmentation and Recovery
• Rotation-Invariant Local-to-Global Representation Learning for 3D Point Cloud
• Fusion 360 Gallery: A Dataset and Environment for Programmatic CAD Construction from Human Design Sequences
• SketchGraphs: A Large-Scale Dataset for Modeling Relational Geometry in Computer-Aided Design
• Materials for Masses: SVBRDF Acquisition with a Single Mobile Phone Image
• EC-Net: an Edge-aware Point set Consolidation Network
• No Shadow Left Behind: Removing Objects and their Shadows using Approximate Lighting and Geometry
• Learning Embedding of 3D models with Quadric Loss
• MRGAN: Multi-Rooted 3D Shape Generation with Unsupervised Part Disentanglement
• GSNet: Joint Vehicle Pose and Shape Reconstruction with Geometrical and Scene-aware Supervision
• GP-Aligner: Unsupervised Non-rigid Groupwise Point Set Registration Based On Optimized Group Latent Descriptor
• Neural Fields in Visual Computing and Beyond
• ForkNet: Multi-branch Volumetric Semantic Completion from a Single Depth Image
• Sparse Tensor-based Multiscale Representation for Point Cloud Geometry Compression
• Nested multi-instance classification
• Learning elementary structures for 3D shape generation and matching
• OperatorNet: Recovering 3D Shapes From Difference Operators
• 3D-R2N2: A Unified Approach for Single and Multi-view 3D Object\n Reconstruction
• SSN: Soft Shadow Network for Image Compositing
• TreeGCN-ED: Encoding Point Cloud using a Tree-Structured Graph Network
• PointGrow: Autoregressively Learned Point Cloud Generation with Self-Attention
• Learning Gradient Fields for Shape Generation
• Sign-Agnostic Implicit Learning of Surface Self-Similarities for Shape Modeling and Reconstruction from Raw Point Clouds
• COALESCE: Component Assembly by Learning to Synthesize Connections
• Learning to Infer Semantic Parameters for 3D Shape Editing
• SnowflakeNet: Point Cloud Completion by Snowflake Point Deconvolution with Skip-Transformer
• QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation
• Self-Supervised Visual Feature Learning With Deep Neural Networks: A Survey
• Beyond Photo Realism for Domain Adaptation from Synthetic Data
• PixelTransformer: Sample Conditioned Signal Generation
• Towards Semantic Segmentation of Urban-Scale 3D Point Clouds: A Dataset, Benchmarks and Challenges
• Multimodal Generative Models for Compositional Representation Learning
• Reconstructing Interactive 3D Scenes by Panoptic Mapping and CAD Model Alignments
• VRGym: A Virtual Testbed for Physical and Interactive AI
• Bundle Optimization for Multi-aspect Embedding
• Rethinking Space-Time Networks with Improved Memory Coverage for\n Efficient Video Object Segmentation
• Geodesic-HOF: 3D Reconstruction Without Cutting Corners
• Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts
• SceneFormer: Indoor Scene Generation with Transformers
• MeshMVS: Multi-View Stereo Guided Mesh Reconstruction
• On the Continuity of Rotation Representations in Neural Networks
• A Review on Deep Learning Techniques Applied to Semantic Segmentation
• Set Distribution Networks: a Generative Model for Sets of Images
• Act the Part: Learning Interaction Strategies for Articulated Object Part Discovery
• 3D Shape Induction from 2D Views of Multiple Objects
• 3D Spatial Recognition without Spatially Labeled 3D
• Syn2Real: A New Benchmark forSynthetic-to-Real Visual Domain Adaptation
• UCLID-Net: Single View Reconstruction in Object Space
• Single-View 3D Object Reconstruction from Shape Priors in Memory
• AccSS3D: Accelerator for Spatially Sparse 3D DNNs
• A Revisit of Shape Editing Techniques: from the Geometric to the Neural Viewpoint
• MirrorNeRF: One-shot Neural Portrait Radiance Field from Multi-mirror Catadioptric Imaging
• PointFlow: 3D Point Cloud Generation with Continuous Normalizing Flows
• Learning 3D Part Assembly from a Single Image
• PU-Transformer: Point Cloud Upsampling Transformer
• SDFDiff: Differentiable Rendering of Signed Distance Fields for 3D Shape Optimization
• From Spectrum Wavelet to Vertex Propagation: Graph Convolutional Networks Based on Taylor Approximation
• 3D Point Cloud Processing and Learning for Autonomous Driving
• Pixels, voxels, and views: A study of shape representations for single view 3D object shape prediction
• Hierarchical Detail Enhancing Mesh-Based Shape Generation with 3D Generative Adversarial Network
• Robust 2D/3D Vehicle Parsing in CVIS
• Physically Realizable Adversarial Examples for LiDAR Object Detection
• 3D-SIS: 3D Semantic Instance Segmentation of RGB-D Scans
• Interpolated Convolutional Networks for 3D Point Cloud Understanding
• Click Here: Human-Localized Keypoints as Guidance for Viewpoint\n Estimation
• PointConv: Deep Convolutional Networks on 3D Point Clouds
• ShapeAdv: Generating Shape-Aware Adversarial 3D Point Clouds
• DIST: Rendering Deep Implicit Signed Distance Function with Differentiable Sphere Tracing
• Latent Transformations for Object View Points Synthesis
• UCSG-Net -- Unsupervised Discovering of Constructive Solid Geometry Tree
• Multiview Based 3D Scene Understanding On Partial Point Sets
• Completing <scp>3D</scp> point clouds of individual trees using deep learning
• Shape Completion using 3D-Encoder-Predictor CNNs and Shape Synthesis
• Convolutional Occupancy Networks
• DualPoseNet: Category-level 6D Object Pose and Size Estimation Using Dual Pose Network with Refined Learning of Pose Consistency
• A Point Set Generation Network for 3D Object Reconstruction from a Single Image
• (AF)2-S3Net: Attentive Feature Fusion with Adaptive Feature Selection for Sparse Semantic Segmentation Network
• RevealNet: Seeing Behind Objects in RGB-D Scans
• Distributional Depth-Based Estimation of Object Articulation Models
• Gotta Adapt 'Em All: Joint Pixel and Feature-Level Domain Adaptation for Recognition in the Wild
• EditVAE: Unsupervised Part-Aware Controllable 3D Point Cloud Shape Generation
• Learning Equivariant Representations
• Cross-modal Center Loss
• 3D Shape Perception from Monocular Vision, Touch, and Shape Priors
• CvxNet: Learnable Convex Decomposition
• Deep Learning for 3D Point Clouds: A Survey
• What Do Single-view 3D Reconstruction Networks Learn?
• Joint 2D-3D-Semantic Data for Indoor Scene Understanding
• 3D Shape Reconstruction from Vision and Touch
• Deformed Implicit Field: Modeling 3D Shapes with Learned Dense Correspondence
• Visual Enhanced 3D Point Cloud Reconstruction from A Single Image
• DA-RNN: Semantic Mapping with Data Associated Recurrent Neural Networks
• Self-Supervised Learning for Domain Adaptation on Point-Clouds
• Learning geometry-image representation for 3D point cloud generation
• Generalization Guarantees for Imitation Learning
• DUDE: Deep Unsigned Distance Embeddings for Hi-Fidelity Representation of Complex 3D Surfaces
• 3D-FUTURE: 3D Furniture shape with TextURE
• A Perceptual Aesthetics Measure for 3D Shapes
• Unsupervised Learning of 3D Structure from Images
• Octree Generating Networks: Efficient Convolutional Architectures for\n High-resolution 3D Outputs
• 3DContextNet: K-d Tree Guided Hierarchical Learning of Point Clouds Using Local and Global Contextual Cues
• Ray-ONet: Efficient 3D Reconstruction From A Single RGB Image
• GraFormer: Graph Convolution Transformer for 3D Pose Estimation
• Deep Learning for 3D Point Cloud Understanding: A Survey
• 3D Point Cloud Denoising using Graph Laplacian Regularization of a Low Dimensional Manifold Model
• Unsupervised Discovery of 3D Physical Objects from Video
• ScanComplete: Large-Scale Scene Completion and Semantic Segmentation for 3D Scans
• TextureNet: Consistent Local Parametrizations for Learning from High-Resolution Signals on Meshes
• UVO Challenge on Video-based Open-World Segmentation 2021: 1st Place Solution
• Skeleton Merger: an Unsupervised Aligned Keypoint Detector
• FlowNet3D: Learning Scene Flow in 3D Point Clouds
• Normalization Techniques in Training DNNs: Methodology, Analysis and Application
• Human Instruction-Following with Deep Reinforcement Learning via Transfer-Learning from Text
• View Synthesis by Appearance Flow
• Deep Mesh Reconstruction from Single RGB Images via Topology Modification Networks
• Inverse Rendering for Complex Indoor Scenes: Shape, Spatially-Varying Lighting and SVBRDF from a Single Image
• Survey: Machine Learning in Production Rendering
• Learning Compositional Representation for 4D Captures with Neural ODE
• CGIntrinsics: Better Intrinsic Image Decomposition through Physically-Based Rendering
• Geo-PIFu: Geometry and Pixel Aligned Implicit Functions for Single-view Human Reconstruction
• Learning to Sit: Synthesizing Human-Chair Interactions via Hierarchical Control
• DeepMVS: Learning Multi-view Stereopsis
• PointContrast: Unsupervised Pre-training for 3D Point Cloud Understanding
• Triplet-Center Loss for Multi-View 3D Object Retrieval
• SpiderCNN: Deep Learning on Point Sets with Parameterized Convolutional Filters
• NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
• Object-Centric Multi-View Aggregation
• Shape from Shading through Shape Evolution
• Hierarchical Surface Prediction for 3D Object Reconstruction
• Generative Models as Distributions of Functions
• Learning to Look Around: Intelligently Exploring Unseen Environments for\n Unknown Tasks
• ApolloCar3D: A Large 3D Car Instance Understanding Benchmark for Autonomous Driving
• Integrating Objects into Monocular SLAM: Line Based Category Specific Models
• SGPN: Similarity Group Proposal Network for 3D Point Cloud Instance Segmentation
• GSPN: Generative Shape Proposal Network for 3D Instance Segmentation in Point Cloud
• Skeleton-bridged Point Completion: From Global Inference to Local Adjustment
• Language Grounding with 3D Objects
• SDM-NET: Deep Generative Network for Structured Deformable Mesh
• ManifoldPlus: A Robust and Scalable Watertight Manifold Surface Generation Method for Triangle Soups
• Learning Deformable Tetrahedral Meshes for 3D Reconstruction
• A Spatial Mapping Algorithm with Applications in Deep Learning-Based\n Structure Classification
• NeRD: Neural 3D Reflection Symmetry Detector
• Predicting Stable Configurations for Semantic Placement of Novel Objects
• Deep Learning for 3D Point Clouds: A Survey
• SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentation
• Diverse Plausible Shape Completions from Ambiguous Depth Images
• pixelNeRF: Neural Radiance Fields from One or Few Images
• RfD-Net: Point Scene Understanding by Semantic Instance Reconstruction
• 3D Point Cloud Denoising via Deep Neural Network based Local Surface Estimation
• Detailed 2D-3D Joint Representation for Human-Object Interaction
• Learning to Synthesize a 4D RGBD Light Field from a Single Image
• Learning Intrinsic Image Decomposition from Watching the World
• Unsupervised Learning of Shape and Pose with Differentiable Point Clouds
• RISA-Net: Rotation-Invariant Structure-Aware Network for Fine-Grained 3D Shape Retrieval
• Deep Differentiable Grasp Planner for High-DOF Grippers
• Fit2Form: 3D Generative Model for Robot Gripper Form Design
• Learning RGB-D Feature Embeddings for Unseen Object Instance Segmentation
• Joint stereo 3D object detection and implicit surface reconstruction
• Self-supervised Single-view 3D Reconstruction via Semantic Consistency
• SeqXY2SeqZ: Structure Learning for 3D Shapes by Sequentially Predicting 1D Occupancy Segments From 2D Coordinates
• Variable-Viewpoint Representations for 3D Object Recognition
• Local Aggressive Adversarial Attacks on 3D Point Cloud
• Rel3D: A Minimally Contrastive Benchmark for Grounding Spatial Relations in 3D
• MOPS-Net: A Matrix Optimization-driven Network forTask-Oriented 3D Point Cloud Downsampling
• FS-Net: Fast Shape-based Network for Category-Level 6D Object Pose Estimation with Decoupled Rotation Mechanism
• Neural Implicit 3D Shapes from Single Images with Spatial Patterns
• Object Parsing in Sequences Using CoordConv Gated Recurrent Networks
• ShapeFlow: Learnable Deformations Among 3D Shapes
• SPSG: Self-Supervised Photometric Scene Generation from RGB-D Scans
• Unsupervised Discovery of Object Radiance Fields
• Do not trust the neighbors! Adversarial Metric Learning for\n Self-Supervised Scene Flow Estimation
• Learning with Algorithmic Supervision via Continuous Relaxations
• Ladybird: Quasi-Monte Carlo Sampling for Deep Implicit Field Based 3D Reconstruction with Symmetry
• SceneNet RGB-D: 5M Photorealistic Images of Synthetic Indoor Trajectories with Ground Truth
• Soft Rasterizer: A Differentiable Renderer for Image-based 3D Reasoning
• NASA: Neural Articulated Shape Approximation
• AtlasNet: A Papier-Mâché Approach to Learning 3D Surface Generation
• Exploit Clues from Views: Self-Supervised and Regularized Learning for Multiview Object Recognition
• Fully Understanding Generic Objects: Modeling, Segmentation, and Reconstruction
• PU-Net: Point Cloud Upsampling Network
• Progressive Point Cloud Deconvolution Generation Network
• Design, Analysis and Application of A Volumetric Convolutional Neural Network
• Normal Assisted Stereo Depth Estimation
• Improved Adversarial Systems for 3D Object Generation and Reconstruction
• DeepSphere: a graph-based spherical CNN
• Grasp Proposal Networks: An End-to-End Solution for Visual Learning of Robotic Grasps
• Im2Struct: Recovering 3D Shape Structure from a Single RGB Image
• Rethinking Loss Design for Large-scale 3D Shape Retrieval
• Semantic Scene Completion from a Single Depth Image
• SPARE3D: A Dataset for SPAtial REasoning on Three-View Line Drawings
• DI-Fusion: Online Implicit 3D Reconstruction with Deep Priors
• Multi-view PointNet for 3D Scene Understanding
• Differentiable Rendering with Perturbed Optimizers
• Learning 3D Dense Correspondence via Canonical Point Autoencoder
• DmifNet:3D Shape Reconstruction Based on Dynamic Multi-Branch Information Fusion
• Self-Supervised Intrinsic Image Decomposition

Related

• ShapeNet: An Information-Rich 3D Model Repository
• 3D ShapeNets: A Deep Representation for Volumetric Shapes
• StructureNet: Hierarchical Graph Networks for 3D Shape Generation
• Volumetric Procedural Models for Shape Representation
• AtlasNet: A Papier-Mâché Approach to Learning 3D Surface Generation