Research map / Point clouds / Segmentation and detection
Semantic segmentation of sparse irregular point clouds for leaf/wood discrimination
Yuchen BAI, Jean-Baptiste Durand, Grégoire Laurent Vincent, Florence Forbes
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 |
|---|---|---|---|
| rebalanced loss (random selection of leaf points) instead of focal loss (ablation) | SOUL · French Guiana ULS | specificity: 0.395 -> 0.631 | Improved |
| rebalanced loss instead of focal loss (ablation) | SOUL · French Guiana ULS | G-mean: 0.615 -> 0.744 | Improved |
| rebalanced loss instead of focal loss (ablation) | SOUL · French Guiana ULS | balanced accuracy: 0.677 -> 0.757 | Improved |
| rebalanced loss instead of focal loss (ablation) | SOUL · French Guiana ULS | accuracy: 0.942 -> 0.826 | Hurt |
| multi-scale geometric features instead of linearity at one scale (ablation) | SOUL · French Guiana ULS | specificity: 0.31 -> 0.631 | Improved |
| multi-scale geometric features instead of linearity at one scale (ablation) | SOUL · French Guiana ULS | balanced accuracy: 0.636 -> 0.757 | Improved |
| multi-scale geometric features instead of linearity at one scale (ablation) | SOUL · French Guiana ULS | accuracy: 0.962 -> 0.826 | Hurt |
| SOUL rebalanced loss vs FSCT | SOUL · French Guiana ULS | specificity: 0.13 -> 0.631 | Improved |
| SOUL rebalanced loss vs LeWos | SOUL · French Guiana ULS | balanced accuracy: 0.520 -> 0.757 | Improved |
| retrain FSCT on ULS data | FSCT · French Guiana ULS | specificity: 0.13 -> 0.01 | Hurt |
| 95% confidence interval over 16 test sets for SOUL | SOUL · French Guiana ULS | specificity: 0.631 -> 0.576 | No clear effect |
| GVD spatial split instead of voxel or sliding-window partition (ablation) | SOUL · French Guiana ULS | border noise | Improved |
Closest papers by meaning
- Bringing SAM to new heights: leveraging elevation data for tree crown segmentation from drone imagery (NeurIPS 2025 · similarity 0.88)
- ForestLPR: LiDAR Place Recognition in Forests Attentioning Multiple BEV Density Images (CVPR 2025 · similarity 0.86)
- Open-Canopy: Towards Very High Resolution Forest Monitoring (CVPR 2025 · similarity 0.84)
- OAM-TCD: A globally diverse dataset of high-resolution tree cover maps (NeurIPS 2024 · similarity 0.84)
- DUNIA: Pixel-Sized Embeddings via Cross-Modal Alignment for Earth Observation Applications (ICML 2025 · similarity 0.84)
- FoMo: Multi-Modal, Multi-Scale and Multi-Task Remote Sensing Foundation Models for Forest Monitoring (AAAI 2025 · similarity 0.84)
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.
