AI Paper Atlas

Research map / Detection and segmentation / CNN features

Interpretation-Oriented Cloud Removal via Observation-Anchored Residual Flow with Geo-Contextual Alignment

Ziyao Wang, Maonan Wang, Yucheng He, Xianping Ma, Ziyi Wang, Hongyang Zhang, Yirong Chen, Man On Pun

ECCV 2026 · Detection and segmentation › CNN features · approach: Domain-specific segmentation and detection · 0 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 MRDM diffusion with OAR-Flow on HDiT (ablation)HDiT backbone · CUHKCR-EXT-CSPSNR (dB): 23.736 -> 23.986Improved
replace MRDM diffusion with OAR-Flow (ablation)HDiT backbone · CUHKCR-EXT-CSSSIM: 0.712 -> 0.698Hurt
add GCPA (GCI loss) to OAR-Flow (ablation)HDiT + OAR-Flow · CUHKCR-EXT-CSPSNR (dB): 23.986 -> 24.230Improved
add GCPA (GCI loss) to OAR-Flow (ablation)HDiT + OAR-Flow · CUHKCR-EXT-CSCLS accuracy: 0.690 -> 0.781Improved
use DiT instead of HDiT with MRDM (ablation)MRDM diffusion · CUHKCR-EXT-CSGFLOPs: 166.72 -> 697.20Hurt
reduce patch size from 2 to 1 (ablation)GACR · Vaihingen-CR-thickPSNR (dB): 33.018 -> 33.799 PSNR; SEG 0.737 -> 0.720; GFLOPs 56.05 -> 223.60Mixed
increase patch size from 2 to 4 (ablation)GACR · Vaihingen-CR-thickPSNR (dB): 33.018 -> 30.303Hurt
vary GCI loss weight lambda away from 0.5 (ablation)GACR · Vaihingen-CR-thickSEG mIoU: 0.737 (lambda 0.5) -> 0.728 (0.2), 0.727 (5.0)Hurt
observation-anchored flow with GCPA vs EMRDM diffusionGACR · Vaihingen-CR-thickPSNR (dB): 28.979 -> 33.018Improved
apply cloud removal before downstream classification instead of training end-to-end on cloudy imagesdownstream CLS · CLS-1 (DINOv3-LVD encoder)accuracy: 0.833 -> 0.833No clear effect
apply GACR cloud removal before segmentationdownstream SEG · SEG-4 (DINOv3-LVD encoder)mIoU: 0.684 -> 0.737Improved
replace DINOv3 LVD downstream weights with SAT weightsdownstream evaluation · CLS-1 (EMRDM-processed)accuracy: 0.776 -> 0.642Hurt

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