ICCV 2025 accepted papers on Point clouds
85 of the 2,701 papers accepted at ICCV 2025 (3.1%) are on Point clouds, compared with 5.2% at ICCV 2023. They fall into 4 topics.
See ICCV 2025 in the interactive map
Submitting to ICCV? Checklist →
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
Point cloud classification · 44 papers
- DriveX: Omni Scene Modeling for Learning Generalizable World Knowledge in Autonomous Driving (14 citations)
- Point Cloud Self-supervised Learning via 3D to Multi-view Masked Learner (13 citations)
- MMGeo: Multimodal Compositional Geo-Localization for UAVs (12 citations)
- Mixed Signals: A Diverse Point Cloud Dataset for Heterogeneous LiDAR V2X Collaboration (10 citations)
- Bridging 3D Anomaly Localization and Repair via High-Quality Continuous Geometric Representation (9 citations)
- CLIP-GS: Unifying Vision-Language Representation with 3D Gaussian Splatting (9 citations)
- Efficient Spiking Point Mamba for Point Cloud Analysis (9 citations)
- PlaceIt3D: Language-Guided Object Placement in Real 3D Scenes (9 citations)
- Towards More Diverse and Challenging Pre-training for Point Cloud Learning: Self-Supervised Cross Reconstruction with Decoupled Views (9 citations)
- Open-Vocabulary Octree-Graph for 3D Scene Understanding (7 citations)
- StruMamba3D: Exploring Structural Mamba for Self-supervised Point Cloud Representation Learning (7 citations)
- CMT: A Cascade MAR with Topology Predictor for Multimodal Conditional CAD Generation (6 citations)
and 32 more in the interactive map.
Shape and surface completion · 17 papers
- CAD-Recode: Reverse Engineering CAD Code from Point Clouds (60 citations)
- HORT: Monocular Hand-held Objects Reconstruction with Transformers (12 citations)
- Learning Normals of Noisy Points by Local Gradient-Aware Surface Filtering (7 citations)
- Revisiting Point Cloud Completion: Are We Ready For The Real-World? (7 citations)
- Noise2Score3D: Tweedie's Approach for Unsupervised Point Cloud Denoising (6 citations)
- Blended Point Cloud Diffusion for Localized Text-guided Shape Editing (5 citations)
- CanFields: Consolidating Diffeomorphic Flows for Non-Rigid 4D Interpolation from Arbitrary-Length Sequences (5 citations)
- Geometric Alignment and Prior Modulation for View-Guided Point Cloud Completion on Unseen Categories (3 citations)
- Guiding Diffusion-Based Articulated Object Generation by Partial Point Cloud Alignment and Physical Plausibility Constraints (3 citations)
- DiffPCI: Large Motion Point Cloud frame Interpolation with Diffusion Model (1 citation)
- Leaps and Bounds: An Improved Point Cloud Winding Number Formulation for Fast Normal Estimation and Surface Reconstruction (1 citation)
- Tree Skeletonization from 3D Point Clouds by Denoising Diffusion (1 citation)
and 5 more in the interactive map.
Segmentation and detection · 13 papers
- ForestFormer3D: A Unified Framework for End-to-End Segmentation of Forest LiDAR 3D Point Clouds (32 citations)
- UAVScenes: A Multi-Modal Dataset for UAVs (22 citations)
- DoppDrive: Doppler-Driven Temporal Aggregation for Improved Radar Object Detection (7 citations)
- All in One: Visual-Description-Guided Unified Point Cloud Segmentation (3 citations)
- CutS3D: Cutting Semantics in 3D for 2D Unsupervised Instance Segmentation (3 citations)
- GeoFormer: Geometry Point Encoder for 3D Object Detection with Graph-based Transformer (3 citations)
- Generalized Few-Shot Point Cloud Segmentation via LLM-Assisted Hyper-Relation Matching (2 citations)
- Robust 3D Object Detection using Probabilistic Point Clouds from Single-Photon LiDARs (2 citations)
- DiffRefine: Diffusion-based Proposal Specific Point Cloud Densification for Cross-Domain Object Detection (1 citation)
- Serialization based Point Cloud Oversegmentation (1 citation)
- SG-LDM: Semantic-Guided LiDAR Generation via Latent-Aligned Diffusion (1 citation)
- Temporal Overlapping Prediction: A Self-supervised Pre-training Method for LiDAR Moving Object Segmentation (1 citation)
and 1 more in the interactive map.
Registration and matching · 11 papers
- TurboReg: TurboClique for Robust and Efficient Point Cloud Registration (24 citations)
- BUFFER-X: Towards Zero-Shot Point Cloud Registration in Diverse Scenes (18 citations)
- TrafficLoc: Localizing Traffic Surveillance Cameras in 3D Scenes (7 citations)
- CA-I2P: Channel-Adaptive Registration Network with Global Optimal Selection (6 citations)
- RARE: Refine Any Registration of Pairwise Point Clouds via Zero-Shot Learning (4 citations)
- Unsupervised RGB-D Point Cloud Registration for Scenes with Low Overlap and Photometric Inconsistency (4 citations)
- HyperGCT: A Dynamic Hyper-GNN-Learned Geometric Constraint for 3D Registration (3 citations)
- Correspondence-Free Fast and Robust Spherical Point Pattern Registration (2 citations)
- ERNet: Efficient Non-Rigid Registration Network for Point Sequences (1 citation)
- MinCD-PnP: Learning 2D-3D Correspondences with Approximate Blind PnP (1 citation)
- Diff2I2P: Differentiable Image-to-Point Cloud Registration with Diffusion Prior
