AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Scaling Up Forest Vision with Synthetic Data

Yihang She, Andrew Blake, David A. Coomes, Srinivasan Keshav

IJCV 2026 · Detection and segmentation › Open-set and few-shot detection · approach: Remote sensing segmentation and change · 5 citations (OpenAlex)

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 changedWhereMeasuredEffect
pretrain on synthetic Sim 11,187 before fine-tuning on 3 coniferous plotsA_simp · Coniferous-NF1 (%): 50.5 -> 79.4Improved
pretrain on synthetic data before fine-tuning on 2 deciduous plotsA_simp · Deciduous-RF1 (%): 31.6 -> 70.0Improved
pretrain on synthetic data before fine-tuning on 2 deciduous plotsA_simp · Deciduous-TF1 (%): 32.7 -> 59.9Improved
fine-tune on a single real plot instead of threeA_simp (pretrained) · Coniferous-NF1 (%): 79.4 -> 77.1Hurt
reduce real training data from all 42 plots to 3 coniferous plots (from scratch)A_simp · Coniferous-NF1 (%): 79.8 -> 50.5Hurt
add physics-based LiDAR simulation instead of mesh nodal points (ablation)A_simp pretraining data · Zero-shot Coniferous-CF1 (%): 5.4 -> 42.6Improved
add scene diversity (4 to 11 scenes) (ablation)A_simp pretraining data · Zero-shot Coniferous-CF1 (%): 42.6 -> 81.5Improved
add more plots per scene (5 to 17) (ablation)A_simp pretraining data · Zero-shot Coniferous-CF1 (%): 81.5 -> 87.0Improved
add scene diversity (4 to 11 scenes) (ablation)A_simp pretraining data · Zero-shot Coniferous-SF1 (%): 52.4 -> 53.0No clear effect
add more plots per scene (5 to 17) (ablation)A_simp pretraining data · Zero-shot Deciduous-RF1 (%): 54.5 -> 61.7Improved
mix synthetic and real labelled data in pretraining (A_full)A_full · Deciduous-Tplot-level F1 (%): 69.4 -> 47.4Hurt
mix synthetic and real labelled data in pretraining (A_full)A_full · Coniferous-Nplot-level F1 (%): 92.4 -> 92.0No clear effect

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy