AI Paper Atlas

Research map / Detection and segmentation / CNN features

Spatial Frequency Modulation for Semantic Segmentation

Linwei Chen, Ying Fu, Lin Gu, Dezhi Zheng, Jifeng Dai

TPAMI 2025 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 45 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 SFMInternImage-T · ADE20K valmIoU: 47.9 -> 49.2 (+1.3)Improved
add SFMInternImage-L · ADE20K valmIoU: 53.9 -> 54.7Improved
add SFMMask2Former Swin-L · ADE20K valmIoU: 56.1 -> 56.7Improved
add SFMMask2Former Swin-T · ADE20K valmIoU: 47.7 -> 49.1 (+1.4)Improved
replace Blur/AdaBlur/FLC low-pass downsampling with SFMUPerNet-ResNet50 · Cityscapes valmIoU: 79.5 vs 78.8, 78.9, 78.6Improved
add ARS to strided-conv downsampling vs DeformConvPCAA-R50 stride 32 · Cityscapes valmIoU: 73.6 -> 74.7Improved
add ARS to DeformConvPCAA-R50 stride 8 · Cityscapes valmIoU: 79.5 -> 80.3Improved
add SFMFCN stride 32 · Cityscapes valmIoU: 71.2 -> 76.2 (+5.0)Improved
add SFM, stride 8 vs stride 32PSPNet · Cityscapes valmIoU: +3.3 at stride 32 (72.7 -> 76.0), +1.1 at stride 8 (78.4 -> 79.5)Mixed
add ARSResNet-50 · CIFAR-100top-1 accuracy: 81.36 -> 82.18Improved
add ARSResNeXt-50 · Tiny-ImageNettop-1 accuracy: 75.39 -> 75.79Improved
add ARS under PGD attackPreact-ResNet-18 FGSM-AT · CIFAR-10PGD robust accuracy: 0.02 -> 32.74Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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