Research map / Point clouds / Shape and surface completion
Syn-to-Real Domain Adaptation for Point Cloud Completion via Part-based Approach
Yunseo Yang, Jihun Kim, Kuk-Jin Yoon
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 refinement module to the part-based baseline (ablation) | part-based completion framework · USSPA | CD x10^2: 9.43 -> 7.38 | Improved |
| add discriminator to the part-based baseline (ablation) | part-based completion framework · USSPA | CD x10^2: 9.43 -> 8.96 | Improved |
| add both discriminator and refinement module (ablation) | part-based completion framework · USSPA | F1-score: 67.4 -> 80.4 | Improved |
| add margin-based filtering of noisy real-domain points (ablation) | part-based decomposition · USSPA | CD x10^2: 8.65 -> 7.51; F1 72.0 -> 77.0 | Improved |
| use part-based decomposition without any discriminator instead of adversarial-only domain adaptation (ablation) | syn-to-real completion · USSPA | CD x10^2: 7.38 (part baseline with refinement, no discriminator) vs 9.01 (USSPA) | Improved |
| use cluster-based co-segmentation for categories lacking part labels | part-based completion · USSPA extra categories | CD x10^2 / F1: 7.86/79.4 vs USSPA 8.36/78.6 | Improved |
| use part-based method on cabinet category | part-based completion · USSPA cabinet | CD x10^2 / F1: 14.03/52.0 vs PoinTr 10.97/64.6 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PoinTr — PoinTr: Diverse Point Cloud Completion With Geometry-Aware Transformers (ICCV 2021) · better on 1 of 1 shared result
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
- PoinTr: Diverse Point Cloud Completion With Geometry-Aware Transformers (ICCV 2021 · similarity 0.89)
- AdaPoinTr: Diverse Point Cloud Completion With Adaptive Geometry-Aware Transformers (TPAMI 2023 · similarity 0.88)
- DAPoinTr: Domain Adaptive Point Transformer for Point Cloud Completion (AAAI 2025 · similarity 0.88)
- Multi-Scale Part-Based Feature Representation for 3D Domain Generalization and Adaptation (TPAMI 2024 · similarity 0.88)
- Optimization over Disentangled Encoding: Unsupervised Cross-Domain Point Cloud Completion via Occlusion Factor Manipulation (ECCV 2022 · similarity 0.88)
- SymmCompletion: High-Fidelity and High-Consistency Point Cloud Completion with Symmetry Guidance (AAAI 2025 · 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.
