AI Paper Atlas

Research map / Detection and segmentation / CNN features

Multi-Scale and Cross-Scale Contrastive Learning for Semantic Segmentation

Theodoros Pissas, Claudio S. Ravasio, Lyndon Da Cruz, Christos Bergeles

ECCV 2022 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 31 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 single-scale contrastive loss (ablation)HRNet HR48v2 with CE · Cityscapes valmIoU: 79.1 -> 80.0Improved
extend contrastive loss to multiple scales (ablation)HRNet HR48v2 · Cityscapes valmIoU: 80.0 -> 80.7Improved
add cross-scale contrastive loss (1 scale pair) to multi-scale loss (ablation)HRNet HR48v2 · Cityscapes valmIoU: 80.7 -> 81.2Improved
use 2 cross-scale pairs instead of 1 (ablation)HRNet HR48v2 · Cityscapes valmIoU: 81.2 -> 81.4Improved
add cross-scale loss to multi-scale loss (ablation)UPerNet Swin-T · ADE20K valmIoU: 45.1 -> 45.8Improved
add multi-scale contrastive loss (ablation)UPerNet Swin-T · ADE20K valmIoU: 44.5 -> 45.1Improved
add balanced anchor sampling instead of using all anchors, single-scale loss (ablation)HRNet · Cityscapes valmIoU: 79.4 -> 80.2Improved
use anchor sampling for multi-scale and cross-scale losses (ablation)HRNet · Cityscapes valmemory: OOM -> 9.7 GB/GPUImproved
use decreasing per-scale weights 1.0 0.7 0.4 0.1 instead of equal weights (ablation)HRNet · Cityscapes valmIoU: 81.8 -> 82.2Improved
apply loss on backbone instead of FPN outputs (ablation)UPerNet Swin-S · ADE20K valmIoU: 49.0 -> 47.9Hurt
add contrastive loss to CEOCRNet ResNet-50 · CaDIS task 3 rare classesIoU: 67.58 -> 73.09Improved
add contrastive loss to CEUPerNet Swin-L · ADE20K val multi-scalemIoU: 53.5 -> 53.3Mixed

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