ICCV 2019 accepted papers on 3D vision
209 of the 1,075 papers accepted at ICCV 2019 (19.4%) are on 3D vision, compared with 20.9% at ICCV 2017. They fall into 5 topics.
See ICCV 2019 in the interactive map
Submitting to ICCV? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Depth and stereo · 68 papers
- Digging Into Self-Supervised Monocular Depth Estimation (2,691 citations)
- Enforcing Geometric Constraints of Virtual Normal for Depth Prediction (518 citations)
- Learning Two-View Correspondences and Geometry Using Order-Aware Network (396 citations)
- Depth From Videos in the Wild: Unsupervised Monocular Depth Learning From Unknown Cameras (391 citations)
- DeepPruner: Learning Efficient Stereo Matching via Differentiable PatchMatch (339 citations)
- Self-Supervised Monocular Depth Hints (286 citations)
- Self-Supervised Learning With Geometric Constraints in Monocular Video: Connecting Flow, Depth, and Camera (270 citations)
- P-MVSNet: Learning Patch-Wise Matching Confidence Aggregation for Multi-View Stereo (269 citations)
- Depth Completion From Sparse LiDAR Data With Depth-Normal Constraints (260 citations)
- Deep Single-Image Portrait Relighting (222 citations)
- Bridging the Domain Gap for Ground-to-Aerial Image Matching (204 citations)
- How Do Neural Networks See Depth in Single Images? (197 citations)
and 56 more in the interactive map.
Human pose estimation · 64 papers
- AMASS: Archive of Motion Capture As Surface Shapes (1,496 citations)
- Learning to Reconstruct 3D Human Pose and Shape via Model-Fitting in the Loop (1,090 citations)
- 6-DOF GraspNet: Variational Grasp Generation for Object Manipulation (613 citations)
- Learning Trajectory Dependencies for Human Motion Prediction (570 citations)
- Pix2Pose: Pixel-Wise Coordinate Regression of Objects for 6D Pose Estimation (546 citations)
- Exploiting Spatial-Temporal Relationships for 3D Pose Estimation via Graph Convolutional Networks (537 citations)
- DPOD: 6D Pose Object Detector and Refiner (513 citations)
- Gaze360: Physically Unconstrained Gaze Estimation in the Wild (508 citations)
- CDPN: Coordinates-Based Disentangled Pose Network for Real-Time RGB-Based 6-DoF Object Pose Estimation (498 citations)
- FreiHAND: A Dataset for Markerless Capture of Hand Pose and Shape From Single RGB Images (480 citations)
- Learnable Triangulation of Human Pose (473 citations)
- Camera Distance-Aware Top-Down Approach for 3D Multi-Person Pose Estimation From a Single RGB Image (363 citations)
and 52 more in the interactive map.
Face and shape · 57 papers
- FSGAN: Subject Agnostic Face Swapping and Reenactment (742 citations)
- Soft Rasterizer: A Differentiable Renderer for Image-Based 3D Reasoning (680 citations)
- Mesh R-CNN (489 citations)
- Multi-Garment Net: Learning to Dress 3D People From Images (446 citations)
- Pix2Vox: Context-Aware 3D Reconstruction From Single and Multi-View Images (384 citations)
- DeepHuman: 3D Human Reconstruction From a Single Image (376 citations)
- HoloGAN: Unsupervised Learning of 3D Representations From Natural Images (350 citations)
- Learning Shape Templates With Structured Implicit Functions (342 citations)
- Tex2Shape: Detailed Full Human Body Geometry From a Single Image (330 citations)
- Texture Fields: Learning Texture Representations in Function Space (266 citations)
- Pixel2Mesh++: Multi-View 3D Mesh Generation via Deformation (251 citations)
- Escaping Plato's Cave: 3D Shape From Adversarial Rendering (218 citations)
and 45 more in the interactive map.
3D generation · 18 papers
- 3D Scene Graph: A Structure for Unified Semantics, 3D Space, and Camera (399 citations)
- Extreme View Synthesis (183 citations)
- Learning Relationships for Multi-View 3D Object Recognition (182 citations)
- 3D Instance Segmentation via Multi-Task Metric Learning (161 citations)
- SceneGraphNet: Neural Message Passing for 3D Indoor Scene Augmentation (85 citations)
- Transformable Bottleneck Networks (82 citations)
- A Dataset of Multi-Illumination Images in the Wild (81 citations)
- Cascaded Context Pyramid for Full-Resolution 3D Semantic Scene Completion (80 citations)
- ForkNet: Multi-Branch Volumetric Semantic Completion From a Single Depth Image (69 citations)
- View N-Gram Network for 3D Object Retrieval (68 citations)
- 3D-RelNet: Joint Object and Relational Network for 3D Prediction (48 citations)
- View Independent Generative Adversarial Network for Novel View Synthesis (36 citations)
and 6 more in the interactive map.
