Research map / Point clouds / Registration and matching
SemReg: Semantics Constrained Point Cloud Registration
Sheldon Fung, Xuequan Lu, Dasith de Silva Edirimuni, Wei Pan, Xiao Liu, HONGDONG LI
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 all four semantic modules (SCM, SAFL, SESA, GGCA) (ablation) | SemReg baseline · 3DLoMatch | RR (%): 66.5 -> 75.9 | Improved |
| add SESA and GGCA to SCM and SAFL (ablation) | SemReg · 3DLoMatch | RR (%): 72.0 -> 75.9 | Improved |
| add Semantics Aware Focal Loss (ablation) | SemReg · 3DLoMatch | RR (%): 75.1 -> 75.9 | Improved |
| add Semantics Constrained Matching (ablation) | SemReg · 3DLoMatch | RR (%): 74.7 -> 75.9 | Improved |
| add SCM alone to the baseline (ablation) | SemReg baseline · 3DLoMatch | RR (%): 66.5 -> 66.9 | Improved |
| add SCM alone to the baseline (ablation) | SemReg baseline · 3DMatch | FMR (%): 97.6 -> 97.3 | Hurt |
| add GMSP guided cross-attention (ablation) | SemReg · 3DLoMatch | RR (%): 73.7 -> 75.9 | Improved |
| add ICP refinement after RANSAC | SemReg · 3DMatch | RR (%): 94.7 -> 95.4 | Improved |
| use SemReg coarse matching versus dense-correspondence methods | SemReg · 3DLoMatch | IR (%): RR 75.9 vs RoITr 74.7; IR 29.4 vs 54.3 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- GeoTransformer — GeoTransformer: Fast and Robust Point Cloud Registration With Geometric Transformer (TPAMI 2023) · better on 3 of 4 shared results
- Lepard — Lepard: Learning Partial Point Cloud Matching in Rigid and Deformable Scenes (CVPR 2022) · better on 2 of 2 shared results
- PCR-CG — PCR-CG: Point Cloud Registration via Deep Explicit Color and Geometry (ECCV 2022) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- ML-SemReg: Boosting Point Cloud Registration with Multi-level Semantic Consistency (ECCV 2024 · similarity 0.93)
- FreeReg: Image-to-Point Cloud Registration Leveraging Pretrained Diffusion Models and Monocular Depth Estimators (ICLR 2024 · similarity 0.91)
- Partial Point Cloud Registration with Multi-view 2D Image Learning (AAAI 2025 · similarity 0.89)
- HybridReg: Robust 3D Point Cloud Registration with Hybrid Motions (AAAI 2025 · similarity 0.88)
- PSReg: Prior-guided Sparse Mixture of Experts for Point Cloud Registration (AAAI 2025 · similarity 0.87)
- Mining and Transferring Feature-Geometry Coherence for Unsupervised Point Cloud Registration (NeurIPS 2024 · similarity 0.87)
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.
