AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

Cross-Patch Dense Contrastive Learning for Semi-Supervised Segmentation of Cellular Nuclei in Histopathologic Images

Huisi Wu, Zhaoze Wang, Youyi Song, Lin Yang, Jing Qin

CVPR 2022 · Label-efficient and robust learning › Semi-supervised learning · approach: Semi-supervised medical segmentation · 89 citations (OpenAlex)

Check 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
Use the full cross-patch dense contrastive framework and auxiliary losses instead of CACDenseUNet (DenseNet-161) semi-supervised segmentation · DSB test, 1/32 labeledDC (%): 86.40 -> 87.49Improved
Use the full cross-patch dense contrastive framework and auxiliary losses instead of CACDenseUNet (DenseNet-161) semi-supervised segmentation · MoNuSeg test, 1/32 labeledDC (%): 74.79 -> 75.97Improved
Use the full cross-patch dense contrastive framework and auxiliary losses instead of CACDenseUNet (DenseNet-161) semi-supervised segmentation · DSB test, 1/16 labeledDC (%): 88.49 -> 89.88Improved
Use the full cross-patch dense contrastive framework and auxiliary losses instead of CACDenseUNet (DenseNet-161) semi-supervised segmentation · MoNuSeg test, 1/16 labeledDC (%): 77.12 -> 77.77Improved
Use the full cross-patch dense contrastive framework and auxiliary losses instead of CACDenseUNet (DenseNet-161) semi-supervised segmentation · DSB test, 1/8 labeledDC (%): 89.37 -> 90.09Improved
Use the full cross-patch dense contrastive framework and auxiliary losses instead of CACDenseUNet (DenseNet-161) semi-supervised segmentation · MoNuSeg test, 1/8 labeledDC (%): 78.24 -> 78.93Improved
Replace random negative sampling with pseudo-label-guided pixel-wise sampling (ablation)Mean-teacher pixel-wise contrastive learning · DSB test, 1/32 labeledDC (%): 81.68 -> 85.54Improved
Replace random negative sampling with pseudo-label-guided pixel-wise sampling (ablation)Mean-teacher pixel-wise contrastive learning · MoNuSeg test, 1/32 labeledDC (%): 69.08 -> 73.69Improved
Add cross-patch pixel- and patch-wise sampling to pseudo-label-guided pixel contrastive learning (ablation)DenseUNet semi-supervised segmentation · DSB test, 1/32 labeledDC (%): 85.54 -> 86.26Improved
Add cross-patch pixel- and patch-wise sampling to pseudo-label-guided pixel contrastive learning (ablation)DenseUNet semi-supervised segmentation · MoNuSeg test, 1/32 labeledDC (%): 73.69 -> 74.49Improved
Add consistency regularization to cross-patch contrastive learning (ablation)DenseUNet semi-supervised segmentation · DSB test, 1/32 labeledDC (%): 86.26 -> 86.83Improved
Add entropy minimization to cross-patch contrastive learning (ablation)DenseUNet semi-supervised segmentation · MoNuSeg test, 1/32 labeledDC (%): 74.49 -> 75.00Improved

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