AI Paper Atlas

Research map / Detection and segmentation / CNN features

PRUE: A Practical Recipe for Field Boundary Segmentation at Scale

Gedeon Muhawenayo, Caleb Robinson, Subash Khanal, Zhanpei Fang, Isaac Corley, Alexander Wollam, Tianyi Gao, Leonard Strnad, Ryan Avery, Lyndon Estes, Ana Tárano, Nathan Jacobs and 1 more

CVPR 2026 · Detection and segmentation › CNN features · approach: Domain-specific segmentation and detection · 6 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
replace the loss with log-cosh Dice (ablation)U-Net EfficientNet-B3 on FTW · FTW testPixel IoU: 0.68 -> 0.77Improved
set boundary class weight to 0.75 (ablation)U-Net EfficientNet-B3 on FTW · FTW testObject F1: 0.39 -> 0.42Improved
use EfficientNet-B7 encoder instead of B3 (ablation)U-Net on FTW · FTW testObject F1: 0.39 -> 0.42Improved
add brightness and resize augmentations (ablation)U-Net EfficientNet-B3 on FTW · FTW testScale sensitivity F1|delta|: 0.15 -> 0.00Improved
add brightness and resize augmentations (ablation)U-Net EfficientNet-B3 on FTW · FTW testObject F1: 0.39 -> 0.38No clear effect
add channel shuffle augmentation (ablation)U-Net EfficientNet-B3 on FTW · FTW testInput-order sensitivity F1|delta|: 0.07 -> 0.00Improved
replace U-Net with FCSiam (ablation)EfficientNet-B3 on FTW · FTW testObject F1: 0.39 -> 0.40No clear effect
combine B7 encoder, log-cosh Dice, weight 0.75, brightness+resize and channel shuffle (ablation)FTW baseline U-Net · FTW testObject F1: 0.38 -> 0.47Improved
use frozen geospatial foundation model encoders with a trained decoder instead of U-NetFTW field boundary segmentation · FTW testPixel IoU: 0.76 -> 0.67 (best GFM, Clay)Hurt
use instance segmentation (zero-shot SAM or Delineate Anything) instead of fine-tuned U-NetFTW field boundary segmentation · FTW testPixel IoU: 0.70 -> 0.32 (SAM zero-shot)Hurt
scale up from U-Net B3 to B7 combinationinference throughput · FTW testThroughput (km2/s): 623.28 -> 306.94Hurt

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