AI Paper Atlas

Research map / Detection and segmentation / CNN features

Enhanced Boundary Learning for Glass-Like Object Segmentation

Hao He, Xiangtai Li, Guangliang Cheng, Jianping Shi, Yunhai Tong, Gaofeng Meng, Véronique Prinet, LuBin Weng

ICCV 2021 · Detection and segmentation › CNN features · approach: Boundary-aware instance segmentation · 127 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 Refined Differential Module (RDM) (ablation)Deeplabv3+ · Trans10k valmIoU: 85.4 -> 89.2Improved
add cascade of RDM modules (ablation)Deeplabv3+ + RDM · Trans10k valmIoU: 89.2 -> 90.1Improved
add Point-based GCN Module (PGM) (ablation)Deeplabv3+ + cascaded RDM · Trans10k valmIoU: 90.1 -> 90.7Improved
add BCE edge loss supervision (ablation)L_merge only · Trans10k valmIoU: 88.3 -> 89.4Improved
add residual-part loss L_residual (ablation)merge + BCE edge loss · Trans10k valmIoU: 89.4 -> 90.4Improved
use Dice loss instead of BCE for edge loss (ablation)EBLNet losses · Trans10k valmIoU: 90.4 -> 90.7Improved
add low-level feature F_low in RDM (ablation)RDM · Trans10k valmIoU: 88.8 -> 89.2Improved
add high-level feature F_high in RDM (ablation)RDM · Trans10k valmIoU: 88.4 -> 89.2Improved
add subtraction and residual learning in RDM (ablation)RDM · Trans10k valmIoU: 86.4 -> 89.2Improved
append RDM+PGM modules to FCNFCN · Trans10k valmIoU: 83.2 -> 87.9Improved
append RDM+PGM modules to PSPNetPSPNet · Trans10k valmIoU: 83.5 -> 88.6Improved
append RDM+PGM modules to DANetDANet · Trans10k valmIoU: 84.1 -> 87.8Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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