Mask2Former instance segmentation model trained on 18,507 high-quality crown annotations from Japanese broadleaf forests achieves strong performance and cross-region generalization to tropical rainforests using only RGB UAV data.
Automatic classification of trees using a uav onboard camera and deep learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.IV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Highly Detailed and Generalizable Broadleaf Tree Crown Instance Segmentation from UAV Imagery
Mask2Former instance segmentation model trained on 18,507 high-quality crown annotations from Japanese broadleaf forests achieves strong performance and cross-region generalization to tropical rainforests using only RGB UAV data.