Research map / Point clouds / Point cloud classification
Fusion-Aware Point Convolution for Online Semantic 3D Scene Segmentation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add fusion-aware online 3D point convolution | Online semantic 3D segmentation · ScanNet | accuracy: 0.566 -> 0.764 | Helped |
| Add fusion-aware online 3D point convolution | Online semantic 3D segmentation · SceneNN | accuracy: 0.666 -> 0.675 | Helped |
| Use the proposed online segmentation method | ScanNet 3D scene segmentation · ScanNet | mean IOU: 0.715 -> 0.720 | Helped |
| Add 2D-3D feature fusion (ablation) | Fusion-aware point convolution · ScanNet | mean IOU: 0.711 -> 0.718 | Helped |
| Add frame-to-frame feature fusion (ablation) | Fusion-aware point convolution · ScanNet | mean IOU: 0.711 -> 0.713 | Helped |
| Add frame-to-frame feature fusion alongside 2D-3D feature fusion (ablation) | Fusion-aware point convolution · ScanNet | mean IOU: 0.718 -> 0.720 | Helped |
| Add 2D-3D feature fusion alongside frame-to-frame feature fusion (ablation) | Fusion-aware point convolution · ScanNet | mean IOU: 0.713 -> 0.720 | Helped |
| Use octree-induced surface-aware neighborhoods instead of Euclidean-distance neighborhoods (ablation) | Point convolution for online segmentation · ScanNet | accuracy | Helped |
| Increase the number of fused scans | Online semantic 3D segmentation · RGB-D sequences | accuracy | Helped |
Closest papers by meaning
- Geometry and Uncertainty-Aware 3D Point Cloud Class-Incremental Semantic Segmentation (CVPR 2023 · similarity 0.87)
- Semantic Segmentation for Real Point Cloud Scenes via Bilateral Augmentation and Adaptive Fusion (CVPR 2021 · similarity 0.87)
- Point Cloud Semantic Scene Completion from RGB-D Images (AAAI 2021 · similarity 0.87)
- Point-MoE: Large-Scale Multi-Dataset Training with Mixture-of-Experts for 3D Semantic Segmentation (ICLR 2026 · similarity 0.87)
- INS-Conv: Incremental Sparse Convolution for Online 3D Segmentation (CVPR 2022 · similarity 0.86)
- Know Your Neighbors: Improving Single-View Reconstruction via Spatial Vision-Language Reasoning (CVPR 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.