Research map / Point clouds / Registration and matching
RCP: Recurrent Closest Point for Point Cloud
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 |
|---|---|---|---|
| Remove point-wise optimization (ablation) | RCP scene-flow method · FlyingThings3D | EPE3D: 0.0403 -> 0.0514 | Hurt |
| Use feature-only point-wise optimization without constraining Z toward X (ablation) | RCP scene-flow method · FlyingThings3D | EPE3D: 0.0403 -> 0.0445 | Hurt |
| Replace softened point-wise selection with Gumbel Softmax during training and argmax at inference (ablation) | RCP scene-flow method · FlyingThings3D | EPE3D: 0.0403 -> 0.0451 | Hurt |
| Use bilateral weights instead of the softened cosine-similarity selection (ablation) | RCP scene-flow method · FlyingThings3D | EPE3D: 0.0403 -> 0.0441 | Hurt |
| Replace recurrent GRU regularization with standard set-convolution layers (ablation) | RCP scene-flow method · FlyingThings3D | EPE3D: 0.0403 -> 0.0491 | Hurt |
| Replace GRU with LSTM recurrent regularization (ablation) | RCP scene-flow method · FlyingThings3D | EPE3D: 0.0403 -> 0.0402 | No clear effect |
| Increase inference iterations beyond the training setting (ablation) | RCP scene-flow method · KITTI and FlyingThings3D | EPE3D: KITTI: 0.0570 -> 0.0481; FlyingThings3D: 0.0403 -> 0.0429 | Mixed |
Closest papers by meaning
- PointFlowNet: Learning Representations for Rigid Motion Estimation From Point Clouds (CVPR 2019 · similarity 0.90)
- Rectified Point Flow: Generic Point Cloud Pose Estimation (NeurIPS 2025 · similarity 0.89)
- FlowNet3D: Learning Scene Flow in 3D Point Clouds (CVPR 2019 · similarity 0.89)
- HCRF-Flow: Scene Flow From Point Clouds With Continuous High-Order CRFs and Position-Aware Flow Embedding (CVPR 2021 · similarity 0.89)
- FlowMamba: Learning Point Cloud Scene Flow with Global Motion Propagation (AAAI 2025 · similarity 0.87)
- ZeroFlow: Scalable Scene Flow via Distillation (ICLR 2024 · similarity 0.86)
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.