ICCV 2021 papers and code focus on point cloud analysis
-
Walk in the Cloud: Learning Curves for Point Clouds Shape Analysis
classification- [Code]
-
Score-Based Point Cloud Denoising
Denoising- [Code]
-
ReDAL: Region-based and Diversity-aware Active Learning for Point Cloud Semantic Segmentation
Segmentation -
HRegNet: A Hierarchical Network for Large-scale Outdoor LiDAR Point Cloud Registration
registration- [Code]
-
Learning with Noisy Labels for Robust Point Cloud Segmentation
segmentationoral- [Code]
-
Sparse-to-dense Feature Matching: Intra and Inter domain Cross-modal Learning in Domain Adaptation for 3D Semantic Segmentation
segmentationDomain Adaptation- [Code]
-
Unsupervised Point Cloud Pre-Training via View-Point Occlusion, Completion
Unsupervised learning- [Code]
-
Group-Free 3D Object Detection via Transformers
3D Object Detection -
Hierarchical Aggregation for 3D Instance Segmentation
segmentation- [Code]
-
3DVG-Transformer: Relation Modeling for Visual Grounding on Point Clouds.
3D Visual Grounding -
Perturbed Self-Distillation: Weakly Supervised Large-Scale Point Cloud Semantic Segmentation.
segmentation -
TempNet: Online Semantic Segmentation on Large-scale Point Cloud Series.
segmentation -
Robustness Certification for Point Cloud Models.
robustness- [Code]
-
Shape Self-Correction for Unsupervised Point Cloud Understanding.
Unsupervised learning -
Pyramid Point Cloud Transformer for Large-Scale Place Recognition.
Place_Recognition- [Code]
-
Geometry-Aware Self-Training for Unsupervised Domain Adaptation on Object Point Clouds.
Domain Adaptation- [Code]
-
Differentiable Convolution Search for Point Cloud Processing.
NAS -
Learning Inner-group Relations on Point Clouds.
classification- [Code]
-
Point-Based Modeling of Human Clothing.
- [Code]
-
Cloud Transformers: A Universal Approach To Point Cloud Processing Tasks.
classification- [Code]