AI Paper Atlas

Research map / Point clouds / Segmentation and detection

SegFly: A 2D-3D-2D Paradigm for Aerial RGB-Thermal Semantic Segmentation at Scale

Markus Gross, Sai Bharadhwaj Matha, Rui Song, Viswanathan Muthuveerappan, Conrad Christoph, Julius Huber, Daniel Cremers

ECCV 2026 · Point clouds › Segmentation and detection · approach: Scene completion and instance segmentation

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
train on SegFly-RGB pseudo-labels instead of manual GT onlyFirefly · SegFly-RGB test, eval on manual GTFWmIoU (%): 45.81 -> 54.90Improved
train on SegFly-RGB pseudo-labels instead of manual GT onlyUPerNet · SegFly-RGB test, eval on manual GTFWmIoU (%): 52.46 -> 54.36Improved
train on SegFly-RGB pseudo-labels instead of manual GT onlySegFormer · SegFly-RGB test, eval on manual GTFWmIoU (%): 48.50 -> 51.36Improved
add RGB pretraining before RGB-T finetuning (ablation)Firefly · SegFly-RGB-T testFWmIoU (%): 35.20 -> 37.38Improved
add RGB-T Rein adaptation stage (ablation)Firefly · SegFly-RGB-T testFWmIoU (%): 37.38 -> 43.83Improved
add RGB-T Rein adaptation stage (ablation)UPerNet · SegFly-RGB-T testAcc. / FWmIoU (%): 33.48 -> 34.74 Acc; 32.21 -> 31.98 FWmIoUMixed
add RGB features by concatenation (RGB-T fusion) (ablation)Firefly · SegFly-RGB-T testAcc./FWmIoU (%): +1.27/+0.43Improved
add occlusion filtering (ablation)point-based rendering · test scene 8accuracy (%): 89.78 -> 90.08Improved
add splatting-based densification (ablation)point-based rendering · test scene 8accuracy (%): 89.54 -> 90.08Improved
add depth-guided label distribution (ablation)point-based rendering · test scene 8accuracy (%): 83.95 -> 90.08Improved
add thermal lens distortion transfer (ablation)RGB-T registration · registration accuracy/FWmIoUaccuracy (%): 84.08 -> 87.05Improved
increase manual annotation from 1.1% to 3.1% (ablation)2D-3D-2D pseudo-labels · test scene 9, RGBaccuracy (%): 73.96 -> 87.53Improved

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