AI Paper Atlas

Research map / Detection and segmentation / CNN features

Piecewise Flat Embedding for Image Segmentation

Chaowei Fang, Zicheng Liao, Yizhou Yu

TPAMI 2018 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation

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 residual-based channel weighting (ablation)soPFE0.8+DB · BSDS500 hierarchical segmentationF_op (ODS): 0.438 -> 0.463Improved
use WSC initialization instead of GMM (with weighting) (ablation)gwPFE0.8+DB · BSDS500 hierarchical segmentationF_op (ODS): 0.440 -> 0.463Improved
vary initialization without channel weighting (GMM, hierarchical, WSC) (ablation)oPFE0.8+DB · BSDS500 hierarchical segmentationF_op (ODS): 0.435, 0.437, 0.438No clear effect
use WSC instead of random initialization with weighting (ablation)wPFE0.8+DB · BSDS500 hierarchical segmentationF_op (ODS): 0.445 -> 0.463Improved
use p=0.8 instead of p=0.5 (ablation)swPFE+DB · BSDS500 hierarchical segmentationF_op (ODS): 0.450 -> 0.463Improved
use p=0.8 instead of p=1 (ablation)swPFE+DB · BSDS500 hierarchical segmentationF_op (ODS): 0.462 -> 0.463No clear effect
replace Laplacian eigenmaps with piecewise flat embeddingclustering segmentation · BSDS500 clustering segmentation (NCut affinity, fixed K)F_op: 54.4% higherImproved
replace eigenmaps with PFE in gPb-owt-ucmgPb-owt-ucm · BSDS500 hierarchical segmentationF_op (ODS): 0.348 -> 0.364Improved
replace eigenmaps with PFE in gPb-owt-ucmgPb-owt-ucm · BSDS500 hierarchical segmentationVI (ODS): 1.690 -> 1.565Improved
replace eigenmaps with PFE in gPb-owt-ucmgPb-owt-ucm · BSDS500 hierarchical segmentationF_b (ODS): 0.729 -> 0.723Hurt
replace Laplacian eigenmaps with PFEDB+eigenmaps · BSDS500 hierarchical segmentationF_op (ODS): 0.445 -> 0.463Improved
replace Laplacian eigenmaps with PFECOB+eigenmaps · PASCAL Context hierarchical segmentationF_op (ODS): 0.350 -> 0.503Improved

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