AIを用いた3Dの研究(ICCV2019)

Total Denoising: Unsupervised Learning of 3D Point Cloud Cleaning

DewarpNet: Single-Image Document Unwarping With Stacked 3D and 2D Regression Networks

Occlusion-Aware Networks for 3D Human Pose Estimation in Video

Disentangling Monocular 3D Object Detection

Moulding Humans: Non-Parametric 3D Human Shape Estimation From Single Images

3DPeople: Modeling the Geometry of Dressed Humans

Learning to Reconstruct 3D Human Pose and Shape via Model-Fitting in the Loop

Exploiting Spatial-Temporal Relationships for 3D Pose Estimation via Graph Convolutional Networks

HEMlets Pose: Learning Part-Centric Heatmap Triplets for Accurate 3D Human Pose Estimation

Pix2Vox: Context-Aware 3D Reconstruction From Single and Multi-View Images

3D-LaneNet: End-to-End 3D Multiple Lane Detection

3D-RelNet: Joint Object and Relational Network for 3D Prediction

DiscoNet: Shapes Learning on Disconnected Manifolds for 3D Editing

Enhancing 2D Representation via Adjacent Views for 3D Shape Retrieval

Few-Shot Generalization for Single-Image 3D Reconstruction via Priors

Unsupervised 3D Reconstruction Networks

Joint Monocular 3D Vehicle Detection and Tracking

PointAE: Point Auto-Encoder for 3D Statistical Shape and Texture Modelling

Multi-Garment Net: Learning to Dress 3D People From Images

Learning an Effective Equivariant 3D Descriptor Without Supervision

MonoLoco: Monocular 3D Pedestrian Localization and Uncertainty Estimation

SO-HandNet: Self-Organizing Network for 3D Hand Pose Estimation With Semi-Supervised Learning

Predicting 3D Human Dynamics From Video

Structured Prediction Helps 3D Human Motion Modelling

Neural 3D Morphable Models: Spiral Convolutional Networks for 3D Shape Representation Learning and Generation

View N-Gram Network for 3D Object Retrieval

HoloGAN: Unsupervised Learning of 3D Representations From Natural Images

FrameNet: Learning Local Canonical Frames of 3D Surfaces From a Single RGB Image

Domain-Adaptive Single-View 3D Reconstruction

RIO: 3D Object Instance Re-Localization in Changing Indoor Environments

Learning to Reconstruct 3D Manhattan Wireframes From a Single Image

Soft Rasterizer: A Differentiable Renderer for Image-Based 3D Reasoning

xR-EgoPose: Egocentric 3D Human Pose From an HMD Camera

DeepHuman: 3D Human Reconstruction From a Single Image

Cascaded Context Pyramid for Full-Resolution 3D Semantic Scene Completion

Asynchronous Single-Photon 3D Imaging

Holistic++ Scene Understanding: Single-View 3D Holistic Scene Parsing and Human Pose Estimation With Human-Object Interaction and Physical Commonsense

Hallucinating IDT Descriptors and I3D Optical Flow Features for Action Recognition With CNNs

GarNet: A Two-Stream Network for Fast and Accurate 3D Cloth Draping

Free-Form Video Inpainting With 3D Gated Convolution and Temporal PatchGAN

Deep Hough Voting for 3D Object Detection in Point Clouds

M3D-RPN: Monocular 3D Region Proposal Network for Object Detection

Face Video Deblurring Using 3D Facial Priors

Learning Joint 2D-3D Representations for Depth Completion

Face De-Occlusion Using 3D Morphable Model and Generative Adversarial Network

Deep Optics for Monocular Depth Estimation and 3D Object Detection