AI Paper Atlas

Research map / Detection and segmentation / CNN features

Scaling Semantic Segmentation Beyond 1K Classes on a Single GPU

Shipra Jain, Danda Pani Paudel, Martin Danelljan, Luc Van Gool

ICCV 2021 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 8 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
sample nearest-neighbour negatives instead of random (ablation)embedding segmentation · Cityscapes (MobileNet)mIoU: 64.49 -> 67.20Improved
add normalization of class embeddings (ablation)embedding segmentation · Cityscapes (MobileNet)mIoU: 67.20 -> 72.56Improved
add max-margin regularization loss (ablation)embedding segmentation · Cityscapes (MobileNet)mIoU: 72.56 -> 73.03Improved
increase number of nearest neighbours k from 4 to 8 (ablation)embedding segmentation · Cityscapes (MobileNet)mIoU: 71.35 -> 71.08No clear effect
increase embedding dimension d from 4 to 7 (ablation)embedding segmentation · CityscapesmIoUImproved
replace d-channel embedding with C-channel outputDeepLabV3+ · COCO+LVIS (1284 classes)mIoU: 1.68 -> 4.57Improved
replace d-channel embedding with C-channel outputDeepLabV3+ · COCO+LVIS (1284 classes)FwIoU: 22.66 -> 39.67Improved
replace d-channel embedding with C-channel outputDeepLabV3+ ResNet50 · ADE20kmIoU: 38.93 -> 38.29No clear effect
replace d-channel embedding with C-channel outputDeepLabV3+ MobileNet · CityscapesmIoU: 72.11 -> 73.03No clear effect
add gradient accumulation (ablation)baseline · COCO+LVISmIoU: 1.68 -> 2.76Improved
add group normalization (ablation)baseline · COCO+LVISmIoU: 1.68 -> 5.15Improved
add group normalization (ablation)ours · COCO+LVISmIoU: 4.57 -> 6.26Improved

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