AAAI 2020 accepted papers on 3D vision
37 of the 1,610 papers accepted at AAAI 2020 (2.3%) are on 3D vision, compared with 2.1% at AAAI 2019. They fall into 4 topics.
See AAAI 2020 in the interactive map
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 · 13 papers
- CSPN++: Learning Context and Resource Aware Convolutional Spatial Propagation Networks for Depth Completion (263 citations)
- Adaptive Unimodal Cost Volume Filtering for Deep Stereo Matching (203 citations)
- Optimal Feature Transport for Cross-View Image Geo-Localization (202 citations)
- AtLoc: Attention Guided Camera Localization (188 citations)
- Attention-Based View Selection Networks for Light-Field Disparity Estimation (156 citations)
- Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume (143 citations)
- Planar Prior Assisted PatchMatch Multi-View Stereo (142 citations)
- Learning to Generate Maps from Trajectories (93 citations)
- Cascading Convolutional Color Constancy (63 citations)
- 3D Crowd Counting via Multi-View Fusion with 3D Gaussian Kernels (40 citations)
- Learning End-to-End Scene Flow by Distilling Single Tasks Knowledge (37 citations)
- Complementary-View Multiple Human Tracking (32 citations)
and 1 more in the interactive map.
Human pose estimation · 13 papers
- 3D Human Pose Estimation Using Spatio-Temporal Networks with Explicit Occlusion Training (146 citations)
- Simple Pose: Rethinking and Improving a Bottom-up Approach for Multi-Person Pose Estimation (106 citations)
- AWR: Adaptive Weighting Regression for 3D Hand Pose Estimation (83 citations)
- FDN: Feature Decoupling Network for Head Pose Estimation (82 citations)
- DGCN: Dynamic Graph Convolutional Network for Efficient Multi-Person Pose Estimation (50 citations)
- Geometry-Driven Self-Supervised Method for 3D Human Pose Estimation (39 citations)
- Kinematic-Structure-Preserved Representation for Unsupervised 3D Human Pose Estimation (38 citations)
- Deep Reinforcement Learning for Active Human Pose Estimation (30 citations)
- 3D Human Pose Estimation via Explicit Compositional Depth Maps (29 citations)
- Chained Representation Cycling: Learning to Estimate 3D Human Pose and Shape by Cycling Between Representations (22 citations)
- Human Synthesis and Scene Compositing (22 citations)
- Progressive Bi-C3D Pose Grammar for Human Pose Estimation (7 citations)
and 1 more in the interactive map.
Face and shape · 8 papers
- LCD: Learned Cross-Domain Descriptors for 2D-3D Matching (89 citations)
- Learning Part Generation and Assembly for Structure-Aware Shape Synthesis (65 citations)
- 3D Shape Completion with Multi-View Consistent Inference (55 citations)
- Realistic Face Reenactment via Self-Supervised Disentangling of Identity and Pose (38 citations)
- Fast and Robust Face-to-Parameter Translation for Game Character Auto-Creation (26 citations)
- GDFace: Gated Deformation for Multi-View Face Image Synthesis (10 citations)
- Geometry-Constrained Car Recognition Using a 3D Perspective Network (10 citations)
- Rank3DGAN: Semantic Mesh Generation Using Relative Attributes (4 citations)
