AI Paper Atlas

Research map / Detection and segmentation / CNN features

EGNet: Edge Guidance Network for Salient Object Detection

Jia-Xing Zhao, Jiang-Jiang Liu, Deng-Ping Fan, Yang Cao, Jufeng Yang, Ming-Ming Cheng

ICCV 2019 · Detection and segmentation › CNN features · approach: RGB-D salient object detection · 1,297 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 salient edge features via side path (edge_PROG) (ablation)U-Net baseline · SODMaxF: .851 -> .873Improved
add salient edge features via side path (edge_PROG) (ablation)U-Net baseline · DUTS-TEMaxF: .855 -> .872Improved
use top-down location propagation for edge features (edge_TDLP) instead of side path (ablation)edge_PROG · SODMaxF: .873 -> .882Improved
use top-down location propagation for edge features (edge_TDLP) instead of side path (ablation)edge_PROG · DUTS-TEMaxF: .872 -> .879Improved
add complementary edge and location modeling (edge_TDLP) (ablation)U-Net baseline · SOD / DUTS-TEMaxF: 3.1%, 2.4%Improved
add NLDF-style IoU edge loss (edge_NLDF) (ablation)U-Net baseline · DUTS-TEMaxF: .855 -> .866No clear effect
add NLDF-style IoU edge loss (edge_NLDF) (ablation)U-Net baseline · SODMaxF: .851 -> .857No clear effect
use one-to-one guidance fusion instead of progressive fusion (ablation)edge_TDLP model · SODMaxF: .882 -> .890Improved
use one-to-one guidance fusion instead of progressive fusion (ablation)edge_TDLP model · DUTS-TEMaxF: .880 -> .893Improved
progressive multi-resolution fusion (MRF_PROG) on top of edge_TDLP (ablation)edge_TDLP model · SODMAE: .100 -> .106Hurt
use EGNet salient edges instead of NLDF edgessalient edge detection · DUTSMaxF: 0.429 -> 0.539Improved
use EGNet instead of PiCANet (ResNet backbone)salient object detection · SODMaxF: 0.861 -> 0.890Improved

Later papers that compared with this method

14 later accepted papers report EGNet as a baseline in their results tables; 10 of them report a better number on most of the results they share with it.

6 more not listed.

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