Papers › Label Decoupling Framework for Salient Object Detection

Label Decoupling Framework for Salient Object Detection

25 Aug 2020CVPR 2020 6arXiv:2008.11048archive 2025-07-28

Jun Wei, Shuhui Wang, Zhe Wu, Chi Su, Qingming Huang, Qi Tian

To get more accurate saliency maps, recent methods mainly focus on aggregating multi-level features from fully convolutional network (FCN) and introducing edge information as auxiliary supervision. Though remarkable progress has been achieved, we observe that the closer the pixel is to the edge, the more difficult it is to be predicted, because edge pixels have a very imbalance distribution. To address this problem, we propose a label decoupling framework (LDF) which consists of a label decoupling (LD) procedure and a feature interaction network (FIN). LD explicitly decomposes the original saliency map into body map and detail map, where body map concentrates on center areas of objects and detail map focuses on regions around edges. Detail map works better because it involves much more pixels than traditional edge supervision. Different from saliency map, body map discards edge pixels and only pays attention to center areas. This successfully avoids the distraction from edge pixels during training. Therefore, we employ two branches in FIN to deal with body map and detail map respectively. Feature interaction (FI) is designed to fuse the two complementary branches to predict the saliency map, which is then used to refine the two branches again. This iterative refinement is helpful for learning better representations and more precise saliency maps. Comprehensive experiments on six benchmark datasets demonstrate that LDF outperforms state-of-the-art approaches on different evaluation metrics.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

weijun88/LDF officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

ObjectObject DetectionRGB Salient Object DetectionSaliency DetectionSalient Object Detectionobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Saliency Detection DUT-OMRON LDF(ours) MAE 0.051 #2 of 5 Archive leaderboard report
Saliency Detection HKU-IS LDF(ours) MAE 0.027 #1 of 3 Archive leaderboard report
Salient Object Detection DUT-OMRON LDF E-measure 0.873 #6 of 8 Archive leaderboard report
Salient Object Detection DUT-OMRON LDF MAE 0.051 #6 of 8 Archive leaderboard report
Salient Object Detection DUT-OMRON LDF S-measure 0.838 #6 of 8 Archive leaderboard report
Salient Object Detection DUT-OMRON LDF max_F1 0.819 #6 of 8 Archive leaderboard report
Salient Object Detection DUTS-TE LDF(ResNet-50) E-measure 0.909 #6 of 8 Archive leaderboard report
Salient Object Detection DUTS-TE LDF(ResNet-50) MAE 0.034 #6 of 8 Archive leaderboard report
Salient Object Detection DUTS-TE LDF(ResNet-50) Smeasure 0.892 #6 of 8 Archive leaderboard report
Salient Object Detection DUTS-TE LDF(ResNet-50) max_F1 0.897 #6 of 8 Archive leaderboard report
Salient Object Detection ECSSD LDF(ours) E-measure 0.924 #8 of 10 Archive leaderboard report
Salient Object Detection ECSSD LDF(ours) MAE 0.033 #8 of 10 Archive leaderboard report
Salient Object Detection ECSSD LDF(ours) S-measure 0.924 #8 of 10 Archive leaderboard report
Salient Object Detection ECSSD LDF(ours) max_F1 0.950 #8 of 10 Archive leaderboard report
Salient Object Detection HKU-IS LDF E-measure 0.953 #6 of 9 Archive leaderboard report
Salient Object Detection HKU-IS LDF MAE 0.027 #6 of 9 Archive leaderboard report
Salient Object Detection HKU-IS LDF S-measure 0.919 #6 of 9 Archive leaderboard report
Salient Object Detection HKU-IS LDF max_F1 0.939 #6 of 9 Archive leaderboard report
Salient Object Detection PASCAL-S LDF(ours) E-measure 0.865 #7 of 10 Archive leaderboard report
Salient Object Detection PASCAL-S LDF(ours) MAE 0.059 #7 of 10 Archive leaderboard report
Salient Object Detection PASCAL-S LDF(ours) S-measure 0.856 #7 of 10 Archive leaderboard report
Salient Object Detection PASCAL-S LDF(ours) max_F1 0.874 #7 of 10 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections