AI Paper Atlas

Research map / Detection and segmentation / CNN features

Learning to Downsample for Segmentation of Ultra-High Resolution Images

Chen Jin, Ryutaro Tanno, Thomy Mertzanidou, Eleftheria Panagiotaki, Daniel C. Alexander

ICLR 2022 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 21 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
Learn sampling locations jointly with segmentation and add edge-loss regularizationDeformation module · Cityscapes, DeepGlobe, and PCa-HistomIoU: 3% to 10% higher absolute mIoU over the "uniform" baselineImproved
Add edge-loss regularization to single segmentation loss (ablation)Deformation module · Cityscapes, DeepGlobe, and PCa-HistomIoUImproved
Use learned non-uniform downsampling instead of uniform downsamplingSegmentation model · Cityscapes fair comparisonmIoU: upto 4% in absolute mIoUImproved
Use learned non-uniform downsampling instead of uniform downsamplingSegmentation model · Cityscapes, DeepGlobe, and PCa-Histocomputational cost at comparable performance: 33%, 90% and 70%Improved
Use single segmentation loss without edge-loss regularization (ablation)Deformation module · Cityscapes, DeepGlobe, and PCa-HistomIoU versus uniform downsampling: 2/3 tested datasetsMixed
Set Gaussian kernel standard deviation from 15 to 26 or 32 (ablation)Deformation module · Cityscapes (1024² to 96² pixels)mIoU: 0.35 -> 0.34; 0.35 -> 0.32Hurt
Set Gaussian kernel standard deviation from 20 to 25 (ablation)Deformation module · DeepGlobe (2448² to 100² pixels)mIoU: 0.43 -> 0.45Improved
Set deformation-module kernel size to 5×5 or 7×7 instead of 3×3 (ablation)Deformation module · Cityscapes (1024² to 96² pixels)mIoU: 0.36 -> 0.35; 0.36 -> 0.35Hurt
Set deformation-module kernel size to 7×7 instead of 3×3 (ablation)Deformation module · DeepGlobe (2448² to 100² pixels)mIoU: 0.45 -> 0.46Improved
Vary deformation-module low-resolution input size (ablation)Deformation module · Cityscapes (1024² to 64² pixels)mIoU: 0.35, 0.35, 0.34, 0.33, 0.36Mixed
Vary deformation-module low-resolution input size (ablation)Deformation module · DeepGlobe (2448² to 50² pixels)mIoU: 0.44, 0.42, 0.45, 0.42, 0.43Mixed
Increase edge-loss weight from γ=50 to γ=100 (ablation)Deformation module · Cityscapes (1024² to 64²)mIoU: 0.33 -> 0.35Improved

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