NeurIPS 2025 accepted papers on Point clouds
48 of the 5,812 papers accepted at NeurIPS 2025 (0.8%) are on Point clouds, compared with 1.1% at NeurIPS 2024. They fall into 4 topics.
See NeurIPS 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 · 28 papers
- Point or Line? Using Line-based Representation for Panoptic Symbol Spotting in CAD Drawings (2 citations)
- Copresheaf Topological Neural Networks: A Generalized Deep Learning Framework (1 citation)
- How Many Tokens Do 3D Point Cloud Transformer Architectures Really Need? (1 citation)
- On Geometry-Enhanced Parameter-Efficient Fine-Tuning for 3D Scene Segmentation (1 citation)
- RIGNO: A Graph-based Framework For Robust And Accurate Operator Learning For PDEs On Arbitrary Domains (1 citation)
- TopoPoint: Enhance Topology Reasoning via Endpoint Detection in Autonomous Driving (1 citation)
- TP-MDDN: Task-Preferenced Multi-Demand-Driven Navigation with Autonomous Decision-Making (1 citation)
- URDF-Anything: Constructing Articulated Objects with 3D Multimodal Language Model (1 citation)
- Asymmetric Dual Self-Distillation for 3D Self-Supervised Representation Learning
- Cloud4D: Estimating Cloud Properties at a High Spatial and Temporal Resolution
- CosmoBench: A Multiscale, Multiview, Multitask Cosmology Benchmark for Geometric Deep Learning
- D$^2$GS: Dense Depth Regularization for LiDAR-free Urban Scene Reconstruction
and 16 more in the interactive map.
Shape and surface completion · 10 papers
- PointMAC: Meta-Learned Adaptation for Robust Test-Time Point Cloud Completion (2 citations)
- APML: Adaptive Probabilistic Matching Loss for Robust 3D Point Cloud Reconstruction (1 citation)
- CAGE: Continuity-Aware edGE Network Unlocks Robust Floorplan Reconstruction (1 citation)
- GeoComplete: Geometry-Aware Diffusion for Reference-Driven Image Completion (1 citation)
- Joint Design of Protein Surface and Backbone Using a Diffusion Bridge Model (1 citation)
- U-CAN: Unsupervised Point Cloud Denoising with Consistency-Aware Noise2Noise Matching (1 citation)
- Complete Structure Guided Point Cloud Completion via Cluster- and Instance-Level Contrastive Learning
- MeshCoder: LLM-Powered Structured Mesh Code Generation from Point Clouds
- MIND: Material Interface Generation from UDFs for Non-Manifold Surface Reconstruction
- Point Cloud Synthesis Using Inner Product Transforms
Segmentation and detection · 6 papers
- SpatialLM: Training Large Language Models for Structured Indoor Modeling (8 citations)
- Reasoning Beyond Points: A Visual Introspective Approach for Few-Shot 3D Segmentation (1 citation)
- TREND: Unsupervised 3D Representation Learning via Temporal Forecasting for LiDAR Perception (1 citation)
- Novel Class Discovery for Point Cloud Segmentation via Joint Learning of Causal Representation and Reasoning
- Point4Bit: Post Training 4-bit Quantization for Point Cloud 3D Detection
- BeyondMix: Leveraging Structural Priors and Long-Range Dependencies for Domain-Invariant LiDAR Segmentation