Cross-Domain Dead Tree Detection via Knowledge Distillation in Aerial Imagery
Researchers have developed a new method for detecting dead trees in aerial imagery using knowledge distillation (KD) to improve model generalization across different forest types. The TreeMort-1T-UNet model, initially trained on Finnish data, was adapted to Polish, German, and Estonian datasets. The feature-level KD approach proved most effective, achieving a Mean Tree IoU of 0.106 and an Instance F1-score of 0.63 on Polish data, while also demonstrating strong representational invariance. AI
IMPACT Enhances transfer learning for remote sensing, offering a scalable tool for ecological monitoring and sustainable forest management.