CVPR 2025 accepted papers on Point clouds
103 of the 2,871 papers accepted at CVPR 2025 (3.6%) are on Point clouds, compared with 3.8% at CVPR 2024. They fall into 4 topics.
See CVPR 2025 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 · 46 papers
- RENO: Real-Time Neural Compression for 3D LiDAR Point Clouds (22 citations)
- Mamba4D: Efficient 4D Point Cloud Video Understanding with Disentangled Spatial-Temporal State Space Models (19 citations)
- Sonata: Self-Supervised Learning of Reliable Point Representations (19 citations)
- Real-IAD D3: A Real-World 2D/Pseudo-3D/3D Dataset for Industrial Anomaly Detection (14 citations)
- TopNet: Transformer-Efficient Occupancy Prediction Network for Octree-Structured Point Cloud Geometry Compression (13 citations)
- Spectral Informed Mamba for Robust Point Cloud Processing (12 citations)
- BWFormer: Building Wireframe Reconstruction from Airborne LiDAR Point Cloud with Transformer (10 citations)
- PO3AD: Predicting Point Offsets toward Better 3D Point Cloud Anomaly Detection (10 citations)
- FG^2: Fine-Grained Cross-View Localization by Fine-Grained Feature Matching (9 citations)
- LidarGait++: Learning Local Features and Size Awareness from LiDAR Point Clouds for 3D Gait Recognition (9 citations)
- Lift3D Policy: Lifting 2D Foundation Models for Robust 3D Robotic Manipulation (9 citations)
- SPARS3R: Semantic Prior Alignment and Regularization for Sparse 3D Reconstruction (9 citations)
and 34 more in the interactive map.
Shape and surface completion · 25 papers
- SPAR3D: Stable Point-Aware Reconstruction of 3D Objects from Single Images (21 citations)
- EdgeDiff: Edge-aware Diffusion Network for Building Reconstruction from Point Clouds (12 citations)
- Scaling Mesh Generation via Compressive Tokenization (9 citations)
- SuperPC: A Single Diffusion Model for Point Cloud Completion, Upsampling, Denoising, and Colorization (9 citations)
- TreeMeshGPT: Artistic Mesh Generation with Autoregressive Tree Sequencing (9 citations)
- A Lightweight UDF Learning Framework for 3D Reconstruction Based on Local Shape Functions (8 citations)
- PCDreamer: Point Cloud Completion Through Multi-view Diffusion Priors (8 citations)
- GenPC: Zero-shot Point Cloud Completion via 3D Generative Priors (7 citations)
- SAMBLE: Shape-Specific Point Cloud Sampling for an Optimal Trade-Off Between Local Detail and Global Uniformity (6 citations)
- High-Fidelity Lightweight Mesh Reconstruction from Point Clouds (5 citations)
- Learning Bijective Surface Parameterization for Inferring Signed Distance Functions from Sparse Point Clouds with Grid Deformation (5 citations)
- 4Deform: Neural Surface Deformation for Robust Shape Interpolation (4 citations)
and 13 more in the interactive map.
Segmentation and detection · 20 papers
- UniMamba: Unified Spatial-Channel Representation Learning with Group-Efficient Mamba for LiDAR-based 3D Object Detection (17 citations)
- An End-to-End Robust Point Cloud Semantic Segmentation Network with Single-Step Conditional Diffusion Models (15 citations)
- BFANet: Revisiting 3D Semantic Segmentation with Boundary Feature Analysis (13 citations)
- Generalized Few-shot 3D Point Cloud Segmentation with Vision-Language Model (12 citations)
- LiMoE: Mixture of LiDAR Representation Learners from Automotive Scenes (9 citations)
- HiLoTs: High-Low Temporal Sensitive Representation Learning for Semi-Supervised LiDAR Segmentation in Autonomous Driving (8 citations)
- Relation3D : Enhancing Relation Modeling for Point Cloud Instance Segmentation (6 citations)
- Multi-Scale Neighborhood Occupancy Masked Autoencoder for Self-Supervised Learning in LiDAR Point Clouds (5 citations)
- Towards Explicit Geometry-Reflectance Collaboration for Generalized LiDAR Segmentation in Adverse Weather (5 citations)
- LogoSP: Local-global Grouping of Superpoints for Unsupervised Semantic Segmentation of 3D Point Clouds (4 citations)
- Masked Point-Entity Contrast for Open-Vocabulary 3D Scene Understanding (4 citations)
- PSA-SSL: Pose and Size-aware Self-Supervised Learning on LiDAR Point Clouds (4 citations)
and 8 more in the interactive map.
Registration and matching · 12 papers
- Dual Focus-Attention Transformer for Robust Point Cloud Registration (15 citations)
- GraphI2P: Image-to-Point Cloud Registration with Exploring Pattern of Correspondence via Graph Learning (8 citations)
- HeMoRa: Unsupervised Heuristic Consensus Sampling for Robust Point Cloud Registration (8 citations)
- Unlocking Generalization Power in LiDAR Point Cloud Registration (6 citations)
- Implicit Correspondence Learning for Image-to-Point Cloud Registration (5 citations)
- Progressive Correspondence Regenerator for Robust 3D Registration (5 citations)
- Zero-shot RGB-D Point Cloud Registration with Pre-trained Large Vision Model (4 citations)
- QuCOOP: A Versatile Framework for Solving Composite and Binary-Parametrised Problems on Quantum Annealers (3 citations)
- Gromov-Wasserstein Problem with Cyclic Symmetry (2 citations)
- AutoURDF: Unsupervised Robot Modeling from Point Cloud Frames Using Cluster Registration (1 citation)
- ColabSfM: Collaborative Structure-from-Motion by Point Cloud Registration (1 citation)
- Practical Solutions to the Relative Pose of Three Calibrated Cameras