Research map / Continual learning / Class-incremental
I3DOL: Incremental 3D Object Learning without Catastrophic Forgetting
Read the paperCheck 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 adaptive-geometric centroid construction (ablation) | I3DOL · ModelNet (increment of 4 classes) | average top-1 accuracy: 81.5 -> 85.3 | Helped |
| add geometric-aware attention (ablation) | I3DOL · ModelNet (increment of 4 classes) | average top-1 accuracy: 83.6 -> 85.3 | Helped |
| add score fairness compensation (ablation) | I3DOL · ModelNet (increment of 4 classes) | average top-1 accuracy: 84.1 -> 85.3 | Helped |
| add adaptive-geometric centroid construction (ablation) | I3DOL · ShapeNet (increment of 6 classes) | average top-1 accuracy: 78.2 -> 81.6 | Helped |
| add geometric-aware attention (ablation) | I3DOL · ShapeNet (increment of 6 classes) | average top-1 accuracy: 79.8 -> 81.6 | Helped |
| add score fairness compensation (ablation) | I3DOL · ShapeNet (increment of 6 classes) | average top-1 accuracy: 80.5 -> 81.6 | Helped |
| add adaptive-geometric centroid construction (ablation) | I3DOL · ScanNet (increment of 2 classes) | average top-1 accuracy: 66.4 -> 70.2 | Helped |
| add geometric-aware attention (ablation) | I3DOL · ScanNet (increment of 2 classes) | average top-1 accuracy: 68.2 -> 70.2 | Helped |
| add score fairness compensation (ablation) | I3DOL · ScanNet (increment of 2 classes) | average top-1 accuracy: 68.7 -> 70.2 | Helped |
Closest papers by meaning
- Activating Sparse Part Concepts for 3D Class Incremental Learning (CVPR 2025 · similarity 0.87)
- MIRACLE 3D: Memory-efficient Integrated Robust Approach for Continual Learning on 3D Point Clouds via Shape Model Construction (ICLR 2025 · similarity 0.83)
- Demystifying Catastrophic Forgetting in Two-Stage Incremental Object Detector (ICML 2025 · similarity 0.81)
- DCA: Dividing and Conquering Amnesia in Incremental Object Detection (AAAI 2025 · similarity 0.80)
- Common3D: Self-Supervised Learning of 3D Morphable Models for Common Objects in Neural Feature Space (CVPR 2025 · similarity 0.80)
- SAR3D: Autoregressive 3D Object Generation and Understanding via Multi-scale 3D VQVAE (CVPR 2025 · similarity 0.80)
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.