CVPR 2023 accepted papers on Point clouds
121 of the 2,359 papers accepted at CVPR 2023 (5.1%) are on Point clouds, compared with 4.7% at CVPR 2022. They fall into 4 topics.
See CVPR 2023 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Point cloud classification · 51 papers
- Multimodal Industrial Anomaly Detection via Hybrid Fusion (211 citations)
- DSVT: Dynamic Sparse Voxel Transformer With Rotated Sets (179 citations)
- Learning 3D Representations From 2D Pre-Trained Models via Image-to-Point Masked Autoencoders (128 citations)
- FlatFormer: Flattened Window Attention for Efficient Point Cloud Transformer (123 citations)
- Self-Positioning Point-Based Transformer for Point Cloud Understanding (123 citations)
- Starting From Non-Parametric Networks for 3D Point Cloud Analysis (121 citations)
- LidarGait: Benchmarking 3D Gait Recognition With Point Clouds (116 citations)
- Meta Architecture for Point Cloud Analysis (113 citations)
- MVImgNet: A Large-Scale Dataset of Multi-View Images (104 citations)
- Attention-Based Point Cloud Edge Sampling (103 citations)
- PointVector: A Vector Representation in Point Cloud Analysis (101 citations)
- CLIP2: Contrastive Language-Image-Point Pretraining From Real-World Point Cloud Data (80 citations)
and 39 more in the interactive map.
Segmentation and detection · 35 papers
- Spherical Transformer for LiDAR-Based 3D Recognition (239 citations)
- PillarNeXt: Rethinking Network Designs for 3D Object Detection in LiDAR Point Clouds (183 citations)
- LoGoNet: Towards Accurate 3D Object Detection With Local-to-Global Cross-Modal Fusion (178 citations)
- CLIP2Scene: Towards Label-Efficient 3D Scene Understanding by CLIP (153 citations)
- RangeViT: Towards Vision Transformers for 3D Semantic Segmentation in Autonomous Driving (138 citations)
- MSeg3D: Multi-Modal 3D Semantic Segmentation for Autonomous Driving (129 citations)
- LaserMix for Semi-Supervised LiDAR Semantic Segmentation (108 citations)
- SCPNet: Semantic Scene Completion on Point Cloud (91 citations)
- PiMAE: Point Cloud and Image Interactive Masked Autoencoders for 3D Object Detection (81 citations)
- PVT-SSD: Single-Stage 3D Object Detector With Point-Voxel Transformer (79 citations)
- ISBNet: A 3D Point Cloud Instance Segmentation Network With Instance-Aware Sampling and Box-Aware Dynamic Convolution (77 citations)
- GrowSP: Unsupervised Semantic Segmentation of 3D Point Clouds (70 citations)
and 23 more in the interactive map.
Shape and surface completion · 26 papers
- Neural Kernel Surface Reconstruction (99 citations)
- Grad-PU: Arbitrary-Scale Point Cloud Upsampling via Gradient Descent With Learned Distance Functions (94 citations)
- ProxyFormer: Proxy Alignment Assisted Point Cloud Completion With Missing Part Sensitive Transformer (93 citations)
- PC2: Projection-Conditioned Point Cloud Diffusion for Single-Image 3D Reconstruction (90 citations)
- AnchorFormer: Point Cloud Completion From Discriminative Nodes (83 citations)
- IterativePFN: True Iterative Point Cloud Filtering (62 citations)
- Towards Better Gradient Consistency for Neural Signed Distance Functions via Level Set Alignment (41 citations)
- HGNet: Learning Hierarchical Geometry From Points, Edges, and Surfaces (38 citations)
- ALTO: Alternating Latent Topologies for Implicit 3D Reconstruction (34 citations)
- Fast Point Cloud Generation With Straight Flows (33 citations)
- Controllable Mesh Generation Through Sparse Latent Point Diffusion Models (30 citations)
- Self-Supervised Learning for Multimodal Non-Rigid 3D Shape Matching (29 citations)
and 14 more in the interactive map.
Registration and matching · 9 papers
- 3D Registration With Maximal Cliques (208 citations)
- Rotation-Invariant Transformer for Point Cloud Matching (143 citations)
- BUFFER: Balancing Accuracy, Efficiency, and Generalizability in Point Cloud Registration (102 citations)
- PEAL: Prior-Embedded Explicit Attention Learning for Low-Overlap Point Cloud Registration (81 citations)
- Unsupervised Deep Probabilistic Approach for Partial Point Cloud Registration (73 citations)
- Robust Multiview Point Cloud Registration With Reliable Pose Graph Initialization and History Reweighting (45 citations)
- Deep Graph-Based Spatial Consistency for Robust Non-Rigid Point Cloud Registration (41 citations)
- Neural Intrinsic Embedding for Non-Rigid Point Cloud Matching (16 citations)
- DeepMapping2: Self-Supervised Large-Scale LiDAR Map Optimization (14 citations)