Research map / Point clouds / Point cloud classification
DG-PIC: Domain Generalized Point-In-Context Learning for Point Cloud Understanding
Jincen Jiang, Qianyu Zhou, Yuhang Li, Xuequan Lu, Meili Wang, Lizhuang Ma, Jian Chang, Jian Jun Zhang
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 |
|---|---|---|---|
| replace random anchor/prototype with global and local prototype-based anchor selection (ablation) | DG-PIC · ScanObjectNN target | Chamfer Distance (reconstruction): 8.4 -> 6.8 | Improved |
| use both global and local prototypes instead of only one level (ablation) | DG-PIC · ScanObjectNN target | Chamfer Distance (denoising): 38.3 / 36.7 -> 35.1 | Improved |
| shift target toward all source domains instead of one nearest domain (uniform averaging) (ablation) | DG-PIC · ScanObjectNN target | Chamfer Distance (reconstruction): 6.8 -> 6.3; registration 6.2 -> 6.5 | No clear effect |
| add macro-level similarity-weighted shifting (ablation) | DG-PIC · ScanObjectNN target | Chamfer Distance (denoising): 32.4 -> 22.7 | Improved |
| add micro-level patch-position shifting (ablation) | DG-PIC · ScanObjectNN target | Chamfer Distance (denoising): 32.4 -> 25.6 | Improved |
| combine macro and micro-level feature shifting (ablation) | DG-PIC · ScanObjectNN target | Chamfer Distance (denoising): 22.7 -> 15.2 | Improved |
| train-time domain generalization mixing (Pointmixup, PointCutMix) in multi-domain multi-task setting | task-specific models · ScanObjectNN target | Chamfer Distance: PointCutMix recon 44.8 vs PointNet 41.1 | No clear effect |
| add test-time dual-level feature shifting and multi-domain prompt pairing to PIC | PIC · ScanObjectNN target | Chamfer Distance (reconstruction): 72.9 -> 4.1 | Improved |
Later papers that compared with this method
1 later accepted paper reports DG-PIC as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Mamba Learns in Context: Structure-Aware Domain Generalization for Multi-Task Point Cloud Understanding (CVPR 2026 · better on 8 of 8 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
- PointDGMamba: Domain Generalization of Point Cloud Classification via Generalized State Space Model (AAAI 2025 · similarity 0.90)
- Point-In-Context: Understanding Point Cloud via In-Context Learning (IJCV 2026 · similarity 0.90)
- Point-PRC: A Prompt Learning Based Regulation Framework for Generalizable Point Cloud Analysis (NeurIPS 2024 · similarity 0.87)
- Construct to Associate: Cooperative Context Learning for Domain Adaptive Point Cloud Segmentation (CVPR 2024 · similarity 0.87)
- Point-Cache: Test-time Dynamic and Hierarchical Cache for Robust and Generalizable Point Cloud Analysis (CVPR 2025 · similarity 0.87)
- Geometrically-driven Aggregation for Zero-shot 3D Point Cloud Understanding (CVPR 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.
