AI Paper Atlas

Research map / Detection and segmentation / Weakly supervised detection and segmentation

Contrastive Learning for Label Efficient Semantic Segmentation

Xiangyun Zhao, Raviteja Vemulapalli, Philip Andrew Mansfield, Boqing Gong, Bradley Green, Lior Shapira, Ying Wu

ICCV 2021 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised semantic segmentation · 195 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 within-image contrastive pretraining before CE fine-tuningDeepLabV3+ · PASCAL VOC 2012 val (2118 images)mIoU: 39.1 -> 62.7Improved
add within-image contrastive pretrainingDeepLabV3+ · Cityscapes val (596 images)mIoU: 63.6 -> 66.9Improved
use cross-image loss instead of within-image loss (ablation)contrastive pretraining · Cityscapes val (596 images)mIoU: 66.9 -> 67.8Improved
use cross-image loss instead of within-image loss (ablation)contrastive pretraining · PASCAL VOC 2012 val (2118 images)mIoU: 62.7 -> 63.5Improved
use batch loss instead of within-image loss (ablation)contrastive pretraining · Cityscapes val (596 images)mIoU: 66.9 -> 66.0Hurt
use batch loss instead of within-image loss (ablation)contrastive pretraining · PASCAL VOC 2012 val (2118 images)mIoU: 62.7 -> 61.7Hurt
pretrain then fine-tune instead of joint training of both losses (ablation)contrastive training strategy · Cityscapes val (596 images)mIoU: 64.6 -> 66.9Improved
pretrain then fine-tune instead of joint training of both losses (ablation)contrastive training strategy · PASCAL VOC 2012 val (2118 images)mIoU: 39.8 -> 62.7Improved
add color distortions in contrastive pretraining (ablation)within-image contrastive pretraining · Cityscapes val (596 images)mIoU: 66.5 -> 67.5Improved
add color distortions in contrastive pretraining (ablation)within-image contrastive pretraining · PASCAL VOC 2012 val (2118 images)mIoU: 62.9 -> 62.7No clear effect
add contrastive pretraining to OCROCR · Cityscapes val (343 images)mIoU: 57.1 -> 63.4Improved
add contrastive pretraining to OCROCR · Cityscapes val (596 images)mIoU: 62.2 -> 65.6Improved

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