The shape variational autoencoder: A deep generative model of part‐segmented 3D objects
Explore this paper's citation graph
Summary
Qualitatively it is demonstrated that the ShapeVAE produces plausible shape samples, and that it captures a semantically meaningful shape‐embedding, and it is shown that the model facilitates mesh reconstruction by sampling consistent surface normals.
- Type
- article
- Published
- 2017-08-01
- Cited by
- 144
- References
- 54
- Access
- Open access
- OpenAlex
- https://openalex.org/W2728326942
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:6054083
Keywords
Computer science, Autoencoder, Embedding, Surface (topology), Artificial intelligence
References
- Deep Boltzmann Machines
- Deep Mixtures of Factor Analysers
- The NURBS Book
- Stochastic Backpropagation and Approximate Inference in Deep Generative Models
- The stitched puppet: A graphical model of 3D human shape and pose
- ShapeSynth: Parameterizing model collections for coupled shape exploration and synthesis
- Auto-Encoding Variational Bayes
- Three-dimensional alpha shapes
- Piecewise smooth surface reconstruction
- Screened poisson surface reconstruction
- Acoustic Modeling Using Deep Belief Networks
- Enriching object detection with 2D-3D registration and continuous viewpoint estimation
- A volumetric method for building complex models from range images
- Smart Variations: Functional Substructures for Part Compatibility
- Estimating image depth using shape collections
- Learning part-based templates from large collections of 3D shapes
- Active Shape Models-Their Training and Application
- Embedded deformation for shape manipulation
- Automatic reconstruction of surfaces and scalar fields from 3D scans
- The Shape Boltzmann Machine: A Strong Model of Object Shape
Cited by
- Learning 3D Shape Completion Under Weak Supervision
- A Survey on Data‐driven Dictionary‐based Methods for 3D Modeling
- Multi-chart generative surface modeling
- Synthesizing cloth wrinkles by CNN‐based geometry image superresolution
- Learning a Representation Map for Robot Navigation using Deep Variational Autoencoder
- Handling Incomplete Heterogeneous Data using VAEs
- Structure-aware Generative Network for 3D-Shape Modeling
- Reconstruction and recommendation of realistic 3D models using cGANs
- Synthesizing Designs With Inter-Part Dependencies Using Hierarchical Generative Adversarial Networks
- Learning to Generate the "Unseen" via Part Synthesis and Composition
- Automatic unpaired shape deformation transfer
- Global-to-local generative model for 3D shapes
- A Novel Variational Autoencoder with Applications to Generative Modelling, Classification, and Ordinal Regression
- GSPN: Generative Shape Proposal Network for 3D Instance Segmentation in Point Cloud
- Composite Shape Modeling via Latent Space Factorization
- Learning Localized Generative Models for 3D Point Clouds via Graph Convolution
- Learning Single-Image 3D Reconstruction by Generative Modelling of Shape, Pose and Shading
- NeuralSampler: Euclidean Point Cloud Auto-Encoder and Sampler
- A generative sampling system for profile designs with shape constraints and user evaluation
- Advances in scene understanding : object detection, reconstruction, layouts, and inference
Related papers
- Three-dimensional Surface Reconstruction Technology in Reverse Engineering
- TC-VAE: Uncovering Out-of-Distribution Data Generative Factors
- Three-dimensional surface reconstruction based on known point cloud normal vector
- Generative Model for Person Re-Identification: A Review
- Sieving Scattered 3D Point Clouds Using Clustering Analysis for 3D Surface Reconstruction
- Multisource Point Clouds, Point Simplification and Surface Reconstruction
- A Complete Method for Reconstructing an Elevation Surface of 3D Point Clouds
- A pipeline for surface reconstruction of 3-dimentional point cloud
- A convolutional autoencoder as a generative model of images for problems of distinguishing attributes and restoring images in missing regions