AI Paper Atlas

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

Weakly-Supervised Semantic Segmentation by Iterative Affinity Learning

Xiang Wang, Sifei Liu, Huimin Ma, Ming-Hsuan Yang

IJCV 2020 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised semantic segmentation · 105 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 mining of confident regions to supervise the pairwise network (ablation)Iterative affinity learning · PASCAL VOC 2012 train (VGG16), step 5mIoU: 56.4 -> 63.1Improved
add mining of confident regions (ablation)Iterative affinity learning · PASCAL VOC 2012 train (VGG16), 500-image energy, step 5energy: 0.048 -> 0.029Improved
add the pairwise affinity network (ablation)Iterative affinity learning · PASCAL VOC 2012 train (VGG16), step 5mIoU: 54.3 -> 63.1Improved
add the pairwise affinity network (ablation)Iterative affinity learning · PASCAL VOC 2012 val (VGG16)mIoU: lower than our method by 9.2%Improved
add region smoothness loss (ablation)Pairwise affinity network · PASCAL VOC 2012 val (VGG16)mIoU: 61.2 -> 62.0Improved
iterate affinity learning for 5 stepsPairwise network · PASCAL VOC 2012 train (VGG16)mIoU: 55.2 -> 63.1Improved
iterate affinity learning for 5 stepsUnary network · PASCAL VOC 2012 train (VGG16)mIoU: 51.5 -> 61.7Improved
iterate beyond step 5Pairwise network · PASCAL VOC 2012 train (VGG16)mIoU: 63.1 -> 63.2No clear effect
refine affinity network with confident regions instead of PSA-style regions, iterativelyIterative PSA · PASCAL VOC 2012 train (VGG16), step 5mIoU: 48.1 -> 63.1Improved
use our confidence-network mining instead of PSA mining for confident-region precisionIterative PSA · PASCAL VOC 2012 train (VGG16), step 5precision: 68.7 -> 81.2Improved
apply the learned pairwise refinement to unary outputUnary network · PASCAL VOC 2012 train (VGG16), step 1mIoU: 51.5 -> 55.2Improved
replace CAM seeds with first-step unary outputCAM seeds · PASCAL VOC 2012 train (VGG16)mIoU: 46.2 -> 51.5Improved

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