AI Paper Atlas

Research map / Detection and segmentation / CNN features

Amulet: Aggregating Multi-Level Convolutional Features for Salient Object Detection

Pingping Zhang, Dong Wang, Huchuan Lu, Hongyu Wang, Xiang Ruan

ICCV 2017 · Detection and segmentation › CNN features · approach: RGB salient object detection · 920 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 the highest-resolution integrated features (Amulet-1/16 to full Amulet) (ablation)Amulet · DUTS-TEF-measure: 0.6891 -> 0.7365Improved
add the highest-resolution integrated features (Amulet-1/16 to full Amulet) (ablation)Amulet · DUTS-TEMAE: 0.09110 -> 0.08517Improved
add the highest-resolution integrated features (Amulet-1/16 to full Amulet) (ablation)Amulet · ECSSDF-measure: 0.8523 -> 0.8684Improved
add the highest-resolution integrated features (Amulet-1/16 to full Amulet) (ablation)Amulet · ECSSDMAE: 0.06477 -> 0.05874Improved
add boundary preserved refinement (BPR) (ablation)Amulet · ECSSDMAE: 0.06572 -> 0.05874Improved
add boundary preserved refinement (BPR) (ablation)Amulet · DUTS-TEMAE: 0.09761 -> 0.08517Improved
add boundary preserved refinement (BPR) (ablation)Amulet · ECSSDF-measure: 0.8647 -> 0.8684Improved
add boundary preserved refinement (BPR) (ablation)Amulet · DUTS-TEF-measure: 0.6912 -> 0.7365Improved
use Amulet-1/16 multi-level aggregation instead of HED aggregationAmulet-1/16 without BPR · ECSSDF-measure: 0.8321 -> 0.8521Improved
use Amulet-1/16 multi-level aggregation instead of FCN-8s aggregationAmulet-1/16 without BPR · ECSSDMAE: 0.1343 -> 0.0662Improved
use Amulet instead of DHSAmulet · PASCAL-SF-measure: 0.7741 -> 0.7632Hurt
use Amulet instead of DCLAmulet · DUT-OMRONF-measure: 0.6842 -> 0.6471Hurt

Later papers that compared with this method

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

3 more not listed.

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