Hybrid active and semi-supervised learning on multimodal satellite and street-view imagery yields an F1-score of 0.90 for urban tree detection with reduced annotation effort.
Health & Place 15, 586–595
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Multimodal Urban Tree Detection from Satellite and Street-Level Imagery via Annotation-Efficient Deep Learning Strategies
Hybrid active and semi-supervised learning on multimodal satellite and street-view imagery yields an F1-score of 0.90 for urban tree detection with reduced annotation effort.