CVPR 2018 accepted papers on 3D vision
183 of the 979 papers accepted at CVPR 2018 (18.7%) are on 3D vision, compared with 17.8% at CVPR 2017. They fall into 4 topics.
See CVPR 2018 in the interactive map
Submitting to CVPR? 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 · 79 papers
- Deep Ordinal Regression Network for Monocular Depth Estimation (1,971 citations)
- Pyramid Stereo Matching Network (1,860 citations)
- CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes (1,709 citations)
- GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose (1,301 citations)
- MegaDepth: Learning Single-View Depth Prediction From Internet Photos (1,151 citations)
- Unsupervised Learning of Depth and Ego-Motion From Monocular Video Using 3D Geometric Constraints (833 citations)
- Benchmarking 6DOF Outdoor Visual Localization in Changing Conditions (720 citations)
- Unsupervised Learning of Monocular Depth Estimation and Visual Odometry With Deep Feature Reconstruction (702 citations)
- Learning Depth From Monocular Videos Using Direct Methods (695 citations)
- Fast and Furious: Real Time End-to-End 3D Detection, Tracking and Motion Forecasting With a Single Convolutional Net (564 citations)
- PAD-Net: Multi-Tasks Guided Prediction-and-Distillation Network for Simultaneous Depth Estimation and Scene Parsing (519 citations)
- DeepMVS: Learning Multi-View Stereopsis (515 citations)
and 67 more in the interactive map.
Human pose estimation · 44 papers
- End-to-End Recovery of Human Shape and Pose (1,958 citations)
- DensePose: Dense Human Pose Estimation in the Wild (1,462 citations)
- Real-Time Seamless Single Shot 6D Object Pose Prediction (943 citations)
- Through-Wall Human Pose Estimation Using Radio Signals (679 citations)
- Learning to Estimate 3D Human Pose and Shape From a Single Color Image (658 citations)
- 2D/3D Pose Estimation and Action Recognition Using Multitask Deep Learning (594 citations)
- A Pose-Sensitive Embedding for Person Re-Identification With Expanded Cross Neighborhood Re-Ranking (555 citations)
- GANerated Hands for Real-Time 3D Hand Tracking From Monocular RGB (555 citations)
- First-Person Hand Action Benchmark With RGB-D Videos and 3D Hand Pose Annotations (526 citations)
- PoseTrack: A Benchmark for Human Pose Estimation and Tracking (516 citations)
- RotationNet: Joint Object Categorization and Pose Estimation Using Multiviews From Unsupervised Viewpoints (506 citations)
- Attention-Aware Compositional Network for Person Re-Identification (493 citations)
and 32 more in the interactive map.
Face and shape · 41 papers
- Neural 3D Mesh Renderer (1,016 citations)
- A Papier-Mâché Approach to Learning 3D Surface Generation (953 citations)
- An Unsupervised Learning Model for Deformable Medical Image Registration (675 citations)
- GVCNN: Group-View Convolutional Neural Networks for 3D Shape Recognition (653 citations)
- Total Capture: A 3D Deformation Model for Tracking Faces, Hands, and Bodies (548 citations)
- Video Based Reconstruction of 3D People Models (496 citations)
- Pix3D: Dataset and Methods for Single-Image 3D Shape Modeling (479 citations)
- Nonlinear 3D Face Morphable Model (423 citations)
- 3D-RCNN: Instance-Level 3D Object Reconstruction via Render-and-Compare (350 citations)
- Unsupervised Training for 3D Morphable Model Regression (336 citations)
- SfSNet: Learning Shape, Reflectance and Illuminance of Faces `in the Wild' (301 citations)
- Self-Supervised Multi-Level Face Model Learning for Monocular Reconstruction at Over 250 Hz (295 citations)
and 29 more in the interactive map.
3D generation · 19 papers
- Triplet-Center Loss for Multi-View 3D Object Retrieval (385 citations)
- ScanComplete: Large-Scale Scene Completion and Semantic Segmentation for 3D Scans (329 citations)
- Viewpoint-Aware Attentive Multi-View Inference for Vehicle Re-Identification (300 citations)
- Semantic Visual Localization (286 citations)
- LayoutNet: Reconstructing the 3D Room Layout From a Single RGB Image (266 citations)
- Learning to Parse Wireframes in Images of Man-Made Environments (240 citations)
- Human-Centric Indoor Scene Synthesis Using Stochastic Grammar (168 citations)
- Lose the Views: Limited Angle CT Reconstruction via Implicit Sinogram Completion (112 citations)
- Good View Hunting: Learning Photo Composition From Dense View Pairs (96 citations)
- Im2Pano3D: Extrapolating 360° Structure and Semantics Beyond the Field of View (74 citations)
- Geometry-Aware Deep Network for Single-Image Novel View Synthesis (66 citations)
- Automatic 3D Indoor Scene Modeling From Single Panorama (59 citations)
and 7 more in the interactive map.
