AI Paper Atlas

Research map / Label-efficient and robust learning / Knowledge distillation

Brewing Stronger Features: Dual-Teacher Distillation for Multispectral Earth Observation

Filip Wolf, Blaž Rolih, Luka Čehovin Zajc

CVPR 2026 · Label-efficient and robust learning › Knowledge distillation · approach: Few-shot and metric distillation · 1 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
add DINOv3 class-token distillation to MS contrastive baseline (ablation)DEO (Swin-Tiny) · segmentation, 3 optical + 3 MSavg mIoU: 69.16 -> 70.94Improved
add separate optical pass and projection head (ablation)DEO (Swin-Tiny) · segmentation, optical avgavg mIoU: 79.07 -> 81.20Improved
add separate optical pass and projection head (ablation)DEO (Swin-Tiny) · segmentation, MS avgavg mIoU: 62.81 -> 62.69Hurt
add DINOv3 patch token distillation (ablation)DEO (Swin-Tiny) · segmentation, 3 optical + 3 MSavg mIoU: optical +0.53; MS 62.69 -> 62.46Mixed
add heavy optical augmentations (ablation)DEO (Swin-Tiny) · segmentation, overallavg mIoU: 72.10 -> 72.48Improved
replace low-res Sentinel-2 optical with high-res aerial imagery in pretraining (ablation)DEO (Swin-Tiny) · segmentation, MS avgavg mIoU: 63.02 -> 63.51Improved
use DINOv3 instead of DINOv2 as distillation teacher (ablation)DEO · segmentation, overallavg mIoU: 73.86 -> 73.90No clear effect
use RADIOv2.5 instead of DINOv3 as teacher (ablation)DEO · segmentation, optical vs MSavg mIoU: optical 85.03 -> 83.57; MS 62.77 -> 63.12Mixed
DEO dual-teacher vs strongest multispectral baselineMS segmentation · multispectral segmentation avgmacro mIoU: 53.25 -> 57.45Improved
DEO vs Scale-MAE on optical change detectionoptical change detection · LEVIRbinary F1: 92.1 -> 91.3Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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