ECCV 2022 accepted papers on Point clouds
99 of the 1,645 papers accepted at ECCV 2022 (6.0%) are on Point clouds, compared with 4.6% at ECCV 2020. They fall into 4 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
Segmentation and detection · 40 papers
- 2DPASS: 2D Priors Assisted Semantic Segmentation on LiDAR Point Clouds (305 citations)
- SQN: Weakly-Supervised Semantic Segmentation of Large-Scale 3D Point Clouds (186 citations)
- CenterFormer: Center-based Transformer for 3D Object Detection (184 citations)
- SWFormer: Sparse Window Transformer for 3D Object Detection in Point Clouds (161 citations)
- FCAF3D: Fully Convolutional Anchor-Free 3D Object Detection (150 citations)
- ProposalContrast: Unsupervised Pre-training for LiDAR-Based 3D Object Detection (110 citations)
- MPPNet: Multi-Frame Feature Intertwining with Proxy Points for 3D Temporal Object Detection (94 citations)
- LiDAR Distillation: Bridging the Beam-Induced Domain Gap for 3D Object Detection (76 citations)
- LESS: Label-Efficient Semantic Segmentation for LiDAR Point Clouds (72 citations)
- Homogeneous Multi-modal Feature Fusion and Interaction for 3D Object Detection (60 citations)
- Lidar Point Cloud Guided Monocular 3D Object Detection (58 citations)
- Salient Object Detection for Point Clouds (54 citations)
and 28 more in the interactive map.
Point cloud classification · 38 papers
- Masked Autoencoders for Point Cloud Self-Supervised Learning (695 citations)
- Masked Discrimination for Self-Supervised Learning on Point Clouds (180 citations)
- "Capturing, Reconstructing, and Simulating: The UrbanScene3D Dataset" (166 citations)
- PointMixer: MLP-Mixer for Point Cloud Understanding (127 citations)
- Bottom Up Top down Detection Transformers for Language Grounding in Images and Point Clouds (105 citations)
- MeshLoc: Mesh-Based Visual Localization (82 citations)
- 3D Siamese Transformer Network for Single Object Tracking on Point Clouds (77 citations)
- Image2Point: 3D Point-Cloud Understanding with 2D Image Pretrained Models (72 citations)
- Bi-PointFlowNet: Bidirectional Learning for Point Cloud Based Scene Flow Estimation (60 citations)
- MeshMAE: Masked Autoencoders for 3D Mesh Data Analysis (54 citations)
- What Matters for 3D Scene Flow Network (49 citations)
- Dynamic 3D Scene Analysis by Point Cloud Accumulation (47 citations)
and 26 more in the interactive map.
Shape and surface completion · 16 papers
- SeedFormer: Patch Seeds Based Point Cloud Completion with Upsample Transformer (222 citations)
- Learning to Generate Realistic LiDAR Point Clouds (92 citations)
- FBNet: Feedback Network for Point Cloud Completion (58 citations)
- PD-Flow: A Point Cloud Denoising Framework with Normalizing Flows (57 citations)
- ExtrudeNet: Unsupervised Inverse Sketch-and-Extrude for Shape Parsing (51 citations)
- Bilateral Normal Integration (49 citations)
- Autoregressive 3D Shape Generation via Canonical Mapping (36 citations)
- SketchSampler: Sketch-Based 3D Reconstruction via View-Dependent Depth Sampling (32 citations)
- Primitive-Based Shape Abstraction via Nonparametric Bayesian Inference (22 citations)
- Few ‘Zero Level Set’-Shot Learning of Shape Signed Distance Functions in Feature Space (19 citations)
- Point Scene Understanding via Disentangled Instance Mesh Reconstruction (19 citations)
- Optimization over Disentangled Encoding: Unsupervised Cross-Domain Point Cloud Completion via Occlusion Factor Manipulation (18 citations)
and 4 more in the interactive map.
Registration and matching · 5 papers
- PCR-CG: Point Cloud Registration via Deep Explicit Color and Geometry (39 citations)
- Improving RGB-D Point Cloud Registration by Learning Multi-Scale Local Linear Transformation (24 citations)
- Learning-Based Point Cloud Registration for 6D Object Pose Estimation in the Real World (24 citations)
- PointCLM: A Contrastive Learning-Based Framework for Multi-Instance Point Cloud Registration (19 citations)
- Semidefinite Relaxations of Truncated Least-Squares in Robust Rotation Search: Tight or Not (4 citations)
