ECCV 2022 accepted papers on 3D vision
306 of the 1,645 papers accepted at ECCV 2022 (18.6%) are on 3D vision, compared with 18.0% at ECCV 2020. They fall into 6 topics.
See ECCV 2022 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
Human pose estimation · 90 papers
- CLIFF: Carrying Location Information in Full Frames into Human Pose and Shape Estimation (309 citations)
- SimCC: A Simple Coordinate Classification Perspective for Human Pose Estimation (291 citations)
- P-STMO: Pre-trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimation (208 citations)
- Gen6D: Generalizable Model-Free 6-DoF Object Pose Estimation from RGB Images (168 citations)
- AvatarPoser: Articulated Full-Body Pose Tracking from Sparse Motion Sensing (151 citations)
- Poseur: Direct Human Pose Regression with Transformers (147 citations)
- Cross-Attention of Disentangled Modalities for 3D Human Mesh Recovery with Transformers (123 citations)
- SmoothNet: A Plug-and-Play Network for Refining Human Poses in Videos (118 citations)
- HuMMan: Multi-modal 4D Human Dataset for Versatile Sensing and Modeling (110 citations)
- DFNet: Enhance Absolute Pose Regression with Direct Feature Matching (104 citations)
- EgoBody: Human Body Shape and Motion of Interacting People from Head-Mounted Devices (104 citations)
- Rethinking Keypoint Representations: Modeling Keypoints and Poses as Objects for Multi-Person Human Pose Estimation (103 citations)
and 78 more in the interactive map.
Face and shape · 83 papers
- Towards Metrical Reconstruction of Human Faces (202 citations)
- SparseNeuS: Fast Generalizable Neural Surface Reconstruction from Sparse Views (170 citations)
- NeuMesh: Learning Disentangled Neural Mesh-Based Implicit Field for Geometry and Texture Editing (143 citations)
- NeuRIS: Neural Reconstruction of Indoor Scenes Using Normal Priors (133 citations)
- TAVA: Template-Free Animatable Volumetric Actors (133 citations)
- 3D Face Reconstruction with Dense Landmarks (126 citations)
- Learning Dynamic Facial Radiance Fields for Few-Shot Talking Head Synthesis (116 citations)
- ARAH: Animatable Volume Rendering of Articulated Human SDFs (113 citations)
- Realistic One-Shot Mesh-Based Head Avatars (99 citations)
- MeshUDF: Fast and Differentiable Meshing of Unsigned Distance Field Networks (95 citations)
- MoFaNeRF: Morphable Facial Neural Radiance Field (89 citations)
- MINER: Multiscale Implicit Neural Representation (63 citations)
and 71 more in the interactive map.
Depth and stereo · 67 papers
- RigNet: Repetitive Image Guided Network for Depth Completion (163 citations)
- MVSTER: Epipolar Transformer for Efficient Multi-View Stereo (158 citations)
- LocalBins: Improving Depth Estimation by Learning Local Distributions (141 citations)
- Secrets of Event-Based Optical Flow (120 citations)
- SimpleRecon: 3D Reconstruction without 3D Convolutions (104 citations)
- Disentangling Object Motion and Occlusion for Unsupervised Multi-Frame Monocular Depth (95 citations)
- RA-Depth: Resolution Adaptive Self-Supervised Monocular Depth Estimation (86 citations)
- Domain Randomization-Enhanced Depth Simulation and Restoration for Perceiving and Grasping Specular and Transparent Objects (83 citations)
- Map-Free Visual Relocalization: Metric Pose Relative to a Single Image (77 citations)
- RelPose: Predicting Probabilistic Relative Rotation for Single Objects in the Wild (77 citations)
- RC-MVSNet: Unsupervised Multi-View Stereo with Neural Rendering (68 citations)
- Is Geometry Enough for Matching in Visual Localization? (57 citations)
and 55 more in the interactive map.
NeRF and radiance fields · 37 papers
- TensoRF: Tensorial Radiance Fields (1,403 citations)
- BungeeNeRF: Progressive Neural Radiance Field for Extreme Multi-Scale Scene Rendering (291 citations)
- NeuMan: Neural Human Radiance Field from a Single Video (244 citations)
- ARF: Artistic Radiance Fields (181 citations)
- SinNeRF: Training Neural Radiance Fields on Complex Scenes from a Single Image (172 citations)
- Semantic-Aware Implicit Neural Audio-Driven Video Portrait Generation (143 citations)
- ActiveNeRF: Learning Where to See with Uncertainty Estimation (133 citations)
- NeRF for Outdoor Scene Relighting (129 citations)
- NeILF: Neural Incident Light Field for Physically-Based Material Estimation (115 citations)
- AdaNeRF: Adaptive Sampling for Real-Time Rendering of Neural Radiance Fields (86 citations)
- Deforming Radiance Fields with Cages (83 citations)
- PS-NeRF: Neural Inverse Rendering for Multi-View Photometric Stereo (83 citations)
and 25 more in the interactive map.
3D generation · 26 papers
- LaMAR: Benchmarking Localization and Mapping for Augmented Reality (110 citations)
- COUCH: Towards Controllable Human-Chair Interactions (107 citations)
- AlignSDF: Pose-Aligned Signed Distance Fields for Hand-Object Reconstruction (73 citations)
- SAGA: Stochastic Whole-Body Grasping with Contact (72 citations)
- TOCH: Spatio-Temporal Object-to-Hand Correspondence for Motion Refinement (70 citations)
- Compositional Human-Scene Interaction Synthesis with Semantic Control (65 citations)
- Grasp’D: Differentiable Contact-Rich Grasp Synthesis for Multi-Fingered Hands (65 citations)
- Learning Ego 3D Representation As Ray Tracing (63 citations)
- InfiniteNature-Zero: Learning Perpetual View Generation of Natural Scenes from Single Images (61 citations)
- ViewFormer: NeRF-Free Neural Rendering from Few Images Using Transformers (59 citations)
- DiffuStereo: High Quality Human Reconstruction via Diffusion-Based Stereo Using Sparse Cameras (57 citations)
- Motion Inspired Unsupervised Perception and Prediction in Autonomous Driving (54 citations)
and 14 more in the interactive map.
