Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling

Abstract

We study the problem of 3D object generation. We propose a novel framework, namely 3D Generative Adversarial Network (3D-GAN), which generates 3D objects from a probabilistic space by leveraging recent advances in volumetric convolutional networks and generative adversarial nets. The benefits of our model are three-fold: first, the use of an adversarial criterion, instead of traditional heuristic criteria, enables the generator to capture object structure implicitly and to synthesize high-quality 3D objects; second, the generator establishes a mapping from a low-dimensional probabilistic space to the space of 3D objects, so that we can sample objects without a reference image or CAD models, and explore the 3D object manifold; third, the adversarial discriminator provides a powerful 3D shape descriptor which, learned without supervision, has wide applications in 3D object recognition. Experiments demonstrate that our method generates high-quality 3D objects, and our unsupervisedly learned features achieve impressive performance on 3D object recognition, comparable with those of supervised learning methods.

arXiv commentNIPS 2016. The first two authors contributed equally to this work

Cited by in corpus (353)

• Generative Adversarial Networks: An Overview
• Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
• Reconstruction of three-dimensional porous media using generative adversarial neural networks
• Image-based 3D Object Reconstruction: State-of-the-Art and Trends in the Deep Learning Era
• How Generative Adversarial Networks and Their Variants Work: An Overview
• Towards Principled Methods for Training Generative Adversarial Networks
• GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis
• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• Mode Regularized Generative Adversarial Networks
• Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitro
• NetGAN: Generating Graphs via Random Walks
• Pix2Vox++: Multi-scale Context-aware 3D Object Reconstruction from Single and Multiple Images
• SegMap: 3D Segment Mapping using Data-Driven Descriptors
• Unsupervised Point Cloud Representation Learning with Deep Neural Networks: A Survey
• Virtual to Real Reinforcement Learning for Autonomous Driving
• Dense 3D Object Reconstruction from a Single Depth View
• 3D Sketching using Multi-View Deep Volumetric Prediction
• Differentiable Rendering: A Survey
• Accelerating 3D Deep Learning with PyTorch3D
• Indoor Scene Understanding in 2.5/3D for Autonomous Agents: A Survey
• Learning to Generate Images of Outdoor Scenes from Attributes and Semantic Layouts
• Geo-PIFu: Geometry and Pixel Aligned Implicit Functions for Single-view Human Reconstruction
• Train Sparsely, Generate Densely: Memory-efficient Unsupervised Training of High-resolution Temporal GAN
• Adaptive O-CNN: A Patch-based Deep Representation of 3D Shapes
• Deep Marching Tetrahedra: a Hybrid Representation for High-Resolution 3D Shape Synthesis
• Unsupervised Learning of 3D Structure from Images
• ComplementMe: Weakly-Supervised Component Suggestions for 3D Modeling
• Self-Contrastive Learning with Hard Negative Sampling for Self-supervised Point Cloud Learning
• Symmetry-adapted generation of 3d point sets for the targeted discovery of molecules
• Neural Marching Cubes
• PsyPhy: A Psychophysics Driven Evaluation Framework for Visual Recognition
• Shape Completion using 3D-Encoder-Predictor CNNs and Shape Synthesis
• Cross-Domain Complementary Learning Using Pose for Multi-Person Part Segmentation
• Learning 3D Shape Completion under Weak Supervision
• Deep Learning for Lung Cancer Detection: Tackling the Kaggle Data Science Bowl 2017 Challenge
• Kaolin: A PyTorch Library for Accelerating 3D Deep Learning Research
• PointFlow: 3D Point Cloud Generation with Continuous Normalizing Flows
• PolyGen: An Autoregressive Generative Model of 3D Meshes
• Improved Adversarial Systems for 3D Object Generation and Reconstruction
• Peeking Behind Objects: Layered Depth Prediction from a Single Image
• GIFS: Neural Implicit Function for General Shape Representation
• Learning Implicit Fields for Generative Shape Modeling
• StructureNet: Hierarchical Graph Networks for 3D Shape Generation
• Category-Level Metric Scale Object Shape and Pose Estimation
• PRS-Net: Planar Reflective Symmetry Detection Net for 3D Models
• Large-Scale 3D Shape Reconstruction and Segmentation from ShapeNet Core55
• Pix3D: Dataset and Methods for Single-Image 3D Shape Modeling
• RotationNet: Joint Object Categorization and Pose Estimation Using Multiviews from Unsupervised Viewpoints
• Unsupervised Generative 3D Shape Learning from Natural Images
• GRAINS: Generative Recursive Autoencoders for INdoor Scenes
• CorrGAN: Sampling Realistic Financial Correlation Matrices Using Generative Adversarial Networks
• Yes, we GAN: Applying Adversarial Techniques for Autonomous Driving
• MeshGAN: Non-linear 3D Morphable Models of Faces
• Self-Supervised Few-Shot Learning on Point Clouds
• A Topology Layer for Machine Learning
• DeepSketchHair: Deep Sketch-based 3D Hair Modeling
• Towards Large-Pose Face Frontalization in the Wild
• A Design Methodology for Efficient Implementation of Deconvolutional Neural Networks on an FPGA
• Rethinking Reprojection: Closing the Loop for Pose-aware ShapeReconstruction from a Single Image
• Neural 3D Morphable Models: Spiral Convolutional Networks for 3D Shape Representation Learning and Generation
• ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing
• BourGAN: Generative Networks with Metric Embeddings
• PF-Net: Point Fractal Network for 3D Point Cloud Completion
• SketchGraphs: A Large-Scale Dataset for Modeling Relational Geometry in Computer-Aided Design
• Text2Shape: Generating Shapes from Natural Language by Learning Joint Embeddings
• Adversarial Generation of Training Examples: Applications to Moving Vehicle License Plate Recognition
• Transformation-Grounded Image Generation Network for Novel 3D View Synthesis
• Dominant Set Clustering and Pooling for Multi-View 3D Object Recognition
• Inverse Graphics GAN: Learning to Generate 3D Shapes from Unstructured 2D Data
• Implicit Geometric Regularization for Learning Shapes
• Triangle-Net: Towards Robustness in Point Cloud Learning
• Single-view 3D Mesh Reconstruction for Seen and Unseen Categories
• 3D-LMNet: Latent Embedding Matching for Accurate and Diverse 3D Point Cloud Reconstruction from a Single Image
• Self-Supervised Deep Learning on Point Clouds by Reconstructing Space
• Neural Unsigned Distance Fields for Implicit Function Learning
• 3D Shape Reconstruction from Sketches via Multi-view Convolutional Networks
• Convolutional Generation of Textured 3D Meshes
• On the Benefit of Adversarial Training for Monocular Depth Estimation
• SDM-NET: Deep Generative Network for Structured Deformable Mesh
• Data-Efficient Learning for Sim-to-Real Robotic Grasping using Deep Point Cloud Prediction Networks
• Nonlinear 3D Cosmic Web Simulation with Heavy-Tailed Generative Adversarial Networks
• Learning to Generate and Reconstruct 3D Meshes with only 2D Supervision
• 3D Object Reconstruction from a Single Depth View with Adversarial Learning
• SurfNet: Generating 3D shape surfaces using deep residual networks
• Adversarial Generation of Continuous Implicit Shape Representations
• High-resolution Deep Convolutional Generative Adversarial Networks
• Data-Driven Approach to Encoding and Decoding 3-D Crystal Structures
• 3D Shape Reconstruction from Vision and Touch
• Deep Learning for LiDAR Point Clouds in Autonomous Driving: A Review
• 3D-A-Nets: 3D Deep Dense Descriptor for Volumetric Shapes with Adversarial Networks
• FoldingNet: Point Cloud Auto-encoder via Deep Grid Deformation
• Shape Inpainting using 3D Generative Adversarial Network and Recurrent Convolutional Networks
• ShapeMOD: Macro Operation Discovery for 3D Shape Programs
• RED: A ReRAM-based Deconvolution Accelerator
• SP-GAN: Sphere-Guided 3D Shape Generation and Manipulation
• Octree Generating Networks: Efficient Convolutional Architectures for High-resolution 3D Outputs
• LPD-Net: 3D Point Cloud Learning for Large-Scale Place Recognition and Environment Analysis
• OpenECAD: An Efficient Visual Language Model for Editable 3D-CAD Design
• Deep Generative Model for Efficient 3D Airfoil Parameterization and Generation
• SoftPool++: An Encoder-Decoder Network for Point Cloud Completion
• Monocular 3D Object Detection Leveraging Accurate Proposals and Shape Reconstruction
• Interactive 3D Modeling with a Generative Adversarial Network
• GRASS: Generative Recursive Autoencoders for Shape Structures
• A Survey on Deep Generative 3D-aware Image Synthesis
• PointGrow: Autoregressively Learned Point Cloud Generation with Self-Attention
• 3D Shape Induction from 2D Views of Multiple Objects
• Cascaded Refinement Network for Point Cloud Completion
• DIST: Rendering Deep Implicit Signed Distance Function with Differentiable Sphere Tracing
• 3DGAUnet: 3D generative adversarial networks with a 3D U-Net based generator to achieve the accurate and effective synthesis of clinical tumor image data for pancreatic cancer
• PPF-FoldNet: Unsupervised Learning of Rotation Invariant 3D Local Descriptors
• Inflating 2D Convolution Weights for Efficient Generation of 3D Medical Images
• Message Passing Multi-Agent GANs
• Fit2Form: 3D Generative Model for Robot Gripper Form Design
• Zero-Shot Text-Guided Object Generation with Dream Fields
• SCORES: Shape Composition with Recursive Substructure Priors
• Unsupervised Learning of Shape and Pose with Differentiable Point Clouds
• Consistent Generative Query Networks
• NeuralSampler: Euclidean Point Cloud Auto-Encoder and Sampler
• CGOF++: Controllable 3D Face Synthesis with Conditional Generative Occupancy Fields
• RGBD-GAN: Unsupervised 3D Representation Learning From Natural Image Datasets via RGBD Image Synthesis
• Learning to Synthesize a 4D RGBD Light Field from a Single Image
• DOVE: Learning Deformable 3D Objects by Watching Videos
• ShapeFlow: Learnable Deformations Among 3D Shapes
• Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wild
• NeuralBlox: Real-Time Neural Representation Fusion for Robust Volumetric Mapping
• Learning Mesh Representations via Binary Space Partitioning Tree Networks
• Learning Pose-invariant 3D Object Reconstruction from Single-view Images
• Convolutional Occupancy Networks
• BSP-Net: Generating Compact Meshes via Binary Space Partitioning
• PU-GAN: a Point Cloud Upsampling Adversarial Network
• Distributional Adversarial Networks
• Self-supervised Modal and View Invariant Feature Learning
• Global-Local Bidirectional Reasoning for Unsupervised Representation Learning of 3D Point Clouds
• Learning Part Generation and Assembly for Structure-aware Shape Synthesis
• Learning Bidirectional LSTM Networks for Synthesizing 3D Mesh Animation Sequences
• Learning Shape Priors for Single-View 3D Completion and Reconstruction
• Blood Vessel Geometry Synthesis using Generative Adversarial Networks
• Early-Phase Performance-Driven Design using Generative Models
• Learning Compositional Radiance Fields of Dynamic Human Heads
• Im2Struct: Recovering 3D Shape Structure from a Single RGB Image
• Neural Radiance Flow for 4D View Synthesis and Video Processing
• How to Hallucinate Functional Proteins
• Improving Semantic Analysis on Point Clouds via Auxiliary Supervision of Local Geometric Priors
• Brick-by-Brick: Combinatorial Construction with Deep Reinforcement Learning
• Conditional Single-view Shape Generation for Multi-view Stereo Reconstruction
• A Survey on Deep Learning Methods for Semantic Image Segmentation in Real-Time
• TM-NET: Deep Generative Networks for Textured Meshes
• Multi-Angle Point Cloud-VAE: Unsupervised Feature Learning for 3D Point Clouds from Multiple Angles by Joint Self-Reconstruction and Half-to-Half Prediction
• Estimating and abstracting the 3D structure of bones using neural networks on X-ray (2D) images
• SingleGAN: Image-to-Image Translation by a Single-Generator Network using Multiple Generative Adversarial Learning
• View Inter-Prediction GAN: Unsupervised Representation Learning for 3D Shapes by Learning Global Shape Memories to Support Local View Predictions
• Learning Gradient Fields for Shape Generation
• 3DViewGraph: Learning Global Features for 3D Shapes from A Graph of Unordered Views with Attention
• Hierarchical Detail Enhancing Mesh-Based Shape Generation with 3D Generative Adversarial Network
• A Skeleton-bridged Deep Learning Approach for Generating Meshes of Complex Topologies from Single RGB Images
• TripletGAN: Training Generative Model with Triplet Loss
• How does Lipschitz Regularization Influence GAN Training?
• Transformable Bottleneck Networks
• Progressive Point Cloud Deconvolution Generation Network
• RfD-Net: Point Scene Understanding by Semantic Instance Reconstruction
• Hybrid Bayesian Eigenobjects: Combining Linear Subspace and Deep Network Methods for 3D Robot Vision
• 3D Shape Perception from Monocular Vision, Touch, and Shape Priors
• Convolutional Neural Networks on non-uniform geometrical signals using Euclidean spectral transformation
• ForkNet: Multi-branch Volumetric Semantic Completion from a Single Depth Image
• Deep Implicit Moving Least-Squares Functions for 3D Reconstruction
• Learning 6-DOF Grasping Interaction via Deep Geometry-aware 3D Representations
• MRI to CT Translation with GANs
• CvxNet: Learnable Convex Decomposition
• PT2PC: Learning to Generate 3D Point Cloud Shapes from Part Tree Conditions
• GestureGAN for Hand Gesture-to-Gesture Translation in the Wild
• Augmenting Implicit Neural Shape Representations with Explicit Deformation Fields
• Spectral-GANs for High-Resolution 3D Point-cloud Generation
• Self-supervised Single-view 3D Reconstruction via Semantic Consistency
• ShapeAdv: Generating Shape-Aware Adversarial 3D Point Clouds
• Parts4Feature: Learning 3D Global Features from Generally Semantic Parts in Multiple Views
• Active Object Reconstruction Using a Guided View Planner
• Weakly Supervised Learning of Multi-Object 3D Scene Decompositions Using Deep Shape Priors
• Combinatorial 3D Shape Generation via Sequential Assembly
• Extreme Relative Pose Estimation for RGB-D Scans via Scene Completion
• Unsupervised Feature Learning for Point Cloud by Contrasting and Clustering With Graph Convolutional Neural Network
• Do 2D GANs Know 3D Shape? Unsupervised 3D shape reconstruction from 2D Image GANs
• DeepPoint: A Deep Learning Model for 3D Reconstruction in Point Clouds via mmWave Radar
• Human-in-the-loop Extraction of Interpretable Concepts in Deep Learning Models
• Hierarchical Point Cloud Encoding and Decoding with Lightweight Self-Attention based Model
• Joint Data and Feature Augmentation for Self-Supervised Representation Learning on Point Clouds
• Self-supervised Feature Learning by Cross-modality and Cross-view Correspondences
• Pix2Shape: Towards Unsupervised Learning of 3D Scenes from Images using a View-based Representation
• Roof-GAN: Learning to Generate Roof Geometry and Relations for Residential Houses
• Unsupervised 3D Learning for Shape Analysis via Multiresolution Instance Discrimination
• COALESCE: Component Assembly by Learning to Synthesize Connections
• 3D Shape Reconstruction from Free-Hand Sketches
• Mask2CAD: 3D Shape Prediction by Learning to Segment and Retrieve
• Semantic Correspondence via 2D-3D-2D Cycle
• Fed-Sim: Federated Simulation for Medical Imaging
• Latent-Space Laplacian Pyramids for Adversarial Representation Learning with 3D Point Clouds
• PQ-NET: A Generative Part Seq2Seq Network for 3D Shapes
• Synthesizing facial photometries and corresponding geometries using generative adversarial networks
• MVPNet: Multi-View Point Regression Networks for 3D Object Reconstruction from A Single Image
• Image-derived generative modeling of pseudo-macromolecular structures - towards the statistical assessment of Electron CryoTomography template matching
• Cerberus: A Multi-headed Derenderer
• FMRI data augmentation via synthesis
• PixelNN: Example-based Image Synthesis
• Deformed Implicit Field: Modeling 3D Shapes with Learned Dense Correspondence
• The Automated Inspection of Opaque Liquid Vaccines
• WarpGAN: Automatic Caricature Generation
• Unsupervised 3D Shape Learning from Image Collections in the Wild
• HyperFlow: Representing 3D Objects as Surfaces
• Feeding the zombies: Synthesizing brain volumes using a 3D progressive growing GAN
• Reconstruction of 3D Porous Media From 2D Slices
• SkeletonNet: A Topology-Preserving Solution for Learning Mesh Reconstruction of Object Surfaces from RGB Images
• Shape As Points: A Differentiable Poisson Solver
• Learning to Transfer: Unsupervised Meta Domain Translation
• Shape Generation using Spatially Partitioned Point Clouds
• A Simple and Scalable Shape Representation for 3D Reconstruction
• TetraTSDF: 3D human reconstruction from a single image with a tetrahedral outer shell
• LP-3DCNN: Unveiling Local Phase in 3D Convolutional Neural Networks
• Local Deep Implicit Functions for 3D Shape
• Unsupervised Partial Point Set Registration via Joint Shape Completion and Registration
• Canonical and Compact Point Cloud Representation for Shape Classification
• SALD: Sign Agnostic Learning with Derivatives
• Hybrid VAE: Improving Deep Generative Models using Partial Observations
• An Acceleration Framework for High Resolution Image Synthesis
• Mesh-based Autoencoders for Localized Deformation Component Analysis
• Influence Estimation for Generative Adversarial Networks
• Correcting differences in multi-site neuroimaging data using Generative Adversarial Networks
• Mesh Variational Autoencoders with Edge Contraction Pooling
• 3D-GMNet: Single-View 3D Shape Recovery as A Gaussian Mixture
• Recursive Variational Autoencoders for 3D Blood Vessel Generative Modeling
• Compressing Representations for Embedded Deep Learning
• DeformSyncNet: Deformation Transfer via Synchronized Shape Deformation Spaces
• Latent feature disentanglement for 3D meshes
• Deep Octree-based CNNs with Output-Guided Skip Connections for 3D Shape and Scene Completion
• 3D Depthwise Convolution: Reducing Model Parameters in 3D Vision Tasks
• Imagine That! Leveraging Emergent Affordances for 3D Tool Synthesis
• Active Perception and Representation for Robotic Manipulation
• GraphTER: Unsupervised Learning of Graph Transformation Equivariant Representations via Auto-Encoding Node-wise Transformations
• Data Augmentation for Enhancing EEG-based Emotion Recognition with Deep Generative Models
• Latent Variable Modeling for Generative Concept Representations and Deep Generative Models
• Volumetric Convolution: Automatic Representation Learning in Unit Ball
• Pipeline Generative Adversarial Networks for Facial Images Generation with Multiple Attributes
• Local Aggressive Adversarial Attacks on 3D Point Cloud
• CodeNeRF: Disentangled Neural Radiance Fields for Object Categories
• Towards Grounding Conceptual Spaces in Neural Representations
• Hypernetwork approach to generating point clouds
• 3D Semantic Scene Completion from a Single Depth Image using Adversarial Training
• A Survey on Deep Geometry Learning: From a Representation Perspective
• CapsGAN: Using Dynamic Routing for Generative Adversarial Networks
• SeqXY2SeqZ: Structure Learning for 3D Shapes by Sequentially Predicting 1D Occupancy Segments From 2D Coordinates
• P${}^2$GNet: Pose-Guided Point Cloud Generating Networks for 6-DoF Object Pose Estimation
• Dynamic Plane Convolutional Occupancy Networks
• GraphX-Convolution for Point Cloud Deformation in 2D-to-3D Conversion
• Training Generative Adversarial Networks with Adaptive Composite Gradient
• Real-time Soft Body 3D Proprioception via Deep Vision-based Sensing
• Semantic Photometric Bundle Adjustment on Natural Sequences
• A Point Cloud Generative Model via Tree-Structured Graph Convolutions for 3D Brain Shape Reconstruction
• HyperCube: Implicit Field Representations of Voxelized 3D Models
• Deep Structure for end-to-end inverse rendering
• Unsupervised Geometric Disentanglement for Surfaces via CFAN-VAE
• Learning a Hierarchical Latent-Variable Model of 3D Shapes
• 3DMaterialGAN: Learning 3D Shape Representation from Latent Space for Materials Science Applications
• From Images to 3D Shape Attributes
• Understanding Pixel-level 2D Image Semantics with 3D Keypoint Knowledge Engine
• How Powerful Are Randomly Initialized Pointcloud Set Functions?
• STD-Net: Structure-preserving and Topology-adaptive Deformation Network for 3D Reconstruction from a Single Image
• Single-View 3D Object Reconstruction from Shape Priors in Memory
• Energy-Based Processes for Exchangeable Data
• Object-Centric Photometric Bundle Adjustment with Deep Shape Prior
• Deriving Visual Semantics from Spatial Context: An Adaptation of LSA and Word2Vec to generate Object and Scene Embeddings from Images
• Learning Permutation Invariant Representations using Memory Networks
• StructEdit: Learning Structural Shape Variations
• Developing Creative AI to Generate Sculptural Objects
• Depth Based Semantic Scene Completion with Position Importance Aware Loss
• RayNet: Learning Volumetric 3D Reconstruction with Ray Potentials
• 3D Brain Reconstruction by Hierarchical Shape-Perception Network from a Single Incomplete Image
• Dense 3D Point Cloud Reconstruction Using a Deep Pyramid Network
• Learning Generative Models of Structured Signals from Their Superposition Using GANs with Application to Denoising and Demixing
• Staying in Shape: Learning Invariant Shape Representations using Contrastive Learning
• A Variational Feature Encoding Method of 3D Object for Probabilistic Semantic SLAM
• Paired 3D Model Generation with Conditional Generative Adversarial Networks
• DECOR-GAN: 3D Shape Detailization by Conditional Refinement
• Diverse Plausible Shape Completions from Ambiguous Depth Images
• DmifNet:3D Shape Reconstruction Based on Dynamic Multi-Branch Information Fusion
• Robust GANs against Dishonest Adversaries
• Learning Embedding of 3D models with Quadric Loss
• Holistic 3D Scene Understanding from a Single Image with Implicit Representation
• Deep Implicit Templates for 3D Shape Representation
• 3D Organ Shape Reconstruction from Topogram Images
• I-nteract 2.0: A Cyber-Physical System to Design 3D Models using Mixed Reality Technologies and Deep Learning for Additive Manufacturing
• Weakly-supervised Generative Adversarial Networks for medical image classification
• Synthesizing 3D Shapes from Silhouette Image Collections using Multi-projection Generative Adversarial Networks
• Inferring 3D Shapes from Image Collections using Adversarial Networks
• Geodesic-HOF: 3D Reconstruction Without Cutting Corners
• A Deep Signed Directional Distance Function for Object Shape Representation
• Learning Quadrangulated Patches For 3D Shape Processing
• Discrete Point Flow Networks for Efficient Point Cloud Generation
• LBS Autoencoder: Self-supervised Fitting of Articulated Meshes to Point Clouds
• Point Discriminative Learning for Data-efficient 3D Point Cloud Analysis
• A curvature and density-based generative representation of shapes
• Learning Single-Image Depth from Videos using Quality Assessment Networks
• Student's t-Generative Adversarial Networks
• MOLTR: Multiple Object Localisation, Tracking, and Reconstruction from Monocular RGB Videos
• D-OccNet: Detailed 3D Reconstruction Using Cross-Domain Learning
• Look, Cast and Mold: Learning 3D Shape Manifold from Single-view Synthetic Data
• Assembling Semantically-Disentangled Representations for Predictive-Generative Models via Adaptation from Synthetic Domain
• Learning to Generate 3D Shapes with Generative Cellular Automata
• Learning geometry-image representation for 3D point cloud generation
• Protecting Anonymous Speech: A Generative Adversarial Network Methodology for Removing Stylistic Indicators in Text
• Weakly-supervised 3D coronary artery reconstruction from two-view angiographic images
• Specular-to-Diffuse Translation for Multi-View Reconstruction
• IVE-GAN: Invariant Encoding Generative Adversarial Networks
• DeepMetaHandles: Learning Deformation Meta-Handles of 3D Meshes with Biharmonic Coordinates
• Scalable Modeling of Spatiotemporal Data using the Variational Autoencoder: an Application in Glaucoma
• Towards a Neural Graphics Pipeline for Controllable Image Generation
• Analytical Derivatives for Differentiable Renderer: 3D Pose Estimation by Silhouette Consistency
• Learning to Reconstruct and Segment 3D Objects
• Neural Implicit 3D Shapes from Single Images with Spatial Patterns
• 3D Topology Transformation with Generative Adversarial Networks
• Synthetic Generation of Three-Dimensional Cancer Cell Models from Histopathological Images
• Doctor of Crosswise: Reducing Over-parametrization in Neural Networks
• Can NetGAN be improved on short random walks?
• Latent Space Exploration Using Generative Kernel PCA
• Not Only Look But Observe: Variational Observation Model of Scene-Level 3D Multi-Object Understanding for Probabilistic SLAM
• ViT-Inception-GAN for Image Colourising
• Shelf-Supervised Mesh Prediction in the Wild
• Unsupervised Learning of Depth and Depth-of-Field Effect from Natural Images with Aperture Rendering Generative Adversarial Networks
• Hallucinating Point Cloud into 3D Sculptural Object
• Inverse Graphics: Unsupervised Learning of 3D Shapes from Single Images
• NeuralQAAD: An Efficient Differentiable Framework for High Resolution Point Cloud Compression
• Point Cloud Colorization Based on Densely Annotated 3D Shape Dataset
• 3D Surface Reconstruction by Pointillism
• Learning Priors for Adversarial Autoencoders
• Image Generation and Recognition (Emotions)
• Towards a Uniform Architecture for the Efficient Implementation of 2D and 3D Deconvolutional Neural Networks on FPGAs
• A Variational Observation Model of 3D Object for Probabilistic Semantic SLAM
• Deep Generative Models: Deterministic Prediction with an Application in Inverse Rendering
• Authoring image decompositions with generative models
• Fast and Explicit Neural View Synthesis
• 3D Shape Generation with Grid-based Implicit Functions
• Improving Model Compatibility of Generative Adversarial Networks by Boundary Calibration
• Profile to Frontal Face Recognition in the Wild Using Coupled Conditional GAN
• Hierarchical View Predictor: Unsupervised 3D Global Feature Learning through Hierarchical Prediction among Unordered Views
• DynOcc: Learning Single-View Depth from Dynamic Occlusion Cues
• 3D Scattering Tomography by Deep Learning with Architecture Tailored to Cloud Fields
• Cascaded Refinement Network for Point Cloud Completion with Self-supervision
• Shape related constraints aware generation of Mechanical Designs through Deep Convolutional GAN
• What Do Single-view 3D Reconstruction Networks Learn?
• 3D-Aware Semantic-Guided Generative Model for Human Synthesis
• Automatic generation of object shapes with desired functionalities
• Design, Analysis and Application of A Volumetric Convolutional Neural Network
• Single Image 3D Object Estimation with Primitive Graph Networks
• 3D-MOV: Audio-Visual LSTM Autoencoder for 3D Reconstruction of Multiple Objects from Video
• GAMesh: Guided and Augmented Meshing for Deep Point Networks
• MV-C3D: A Spatial Correlated Multi-View 3D Convolutional Neural Networks