Papers › Suppress and Balance: A Simple Gated Network for Salient Object Detection

Suppress and Balance: A Simple Gated Network for Salient Object Detection

16 Jul 2020ECCV 2020 8arXiv:2007.08074archive 2025-07-28

Xiaoqi Zhao, Youwei Pang, Lihe Zhang, Huchuan Lu, Lei Zhang

Most salient object detection approaches use U-Net or feature pyramid networks (FPN) as their basic structures. These methods ignore two key problems when the encoder exchanges information with the decoder: one is the lack of interference control between them, the other is without considering the disparity of the contributions of different encoder blocks. In this work, we propose a simple gated network (GateNet) to solve both issues at once. With the help of multilevel gate units, the valuable context information from the encoder can be optimally transmitted to the decoder. We design a novel gated dual branch structure to build the cooperation among different levels of features and improve the discriminability of the whole network. Through the dual branch design, more details of the saliency map can be further restored. In addition, we adopt the atrous spatial pyramid pooling based on the proposed "Fold" operation (Fold-ASPP) to accurately localize salient objects of various scales. Extensive experiments on five challenging datasets demonstrate that the proposed model performs favorably against most state-of-the-art methods under 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

Xiaoqi-Zhao-DLUT/GateNet-RGB-Saliency officialmentioned in papermentioned on GitHubpytorch report
xiaoqi-zhao-dlut/msnet mentioned on GitHubpytorch report
xiaoqi-zhao-dlut/multi-source-aps-zvos mentioned 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

DecoderDichotomous Image SegmentationObject DetectionRGB Salient Object DetectionSalient Object Detectionobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Dichotomous Image Segmentation DIS-TE1 GateNet E-measure 0.766 #19 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE1 GateNet HCE 230 #19 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE1 GateNet MAE 0.099 #19 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE1 GateNet S-Measure 0.701 #19 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE1 GateNet max F-Measure 0.620 #19 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE1 GateNet weighted F-measure 0.517 #19 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE2 GateNet E-measure 0.804 #18 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE2 GateNet HCE 501 #18 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE2 GateNet MAE 0.102 #18 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE2 GateNet S-Measure 0.737 #18 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE2 GateNet max F-Measure 0.702 #18 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE2 GateNet weighted F-measure 0.598 #18 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE3 GateNet E-measure 0.815 #19 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE3 GateNet HCE 972 #19 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE3 GateNet MAE 0.103 #19 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE3 GateNet S-Measure 0.747 #19 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE3 GateNet max F-Measure 0.726 #19 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE3 GateNet weighted F-measure 0.620 #19 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE4 GateNet E-measure 0.803 #16 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE4 GateNet HCE 3654 #16 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE4 GateNet MAE 0.109 #16 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE4 GateNet S-Measure 0.743 #16 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE4 GateNet max F-Measure 0.729 #16 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-TE4 GateNet weighted F-measure 0.625 #16 of 22 Archive leaderboard report
Dichotomous Image Segmentation DIS-VD GateNet E-measure 0.783 #20 of 24 Archive leaderboard report
Dichotomous Image Segmentation DIS-VD GateNet HCE 1493 #20 of 24 Archive leaderboard report
Dichotomous Image Segmentation DIS-VD GateNet MAE 0.110 #20 of 24 Archive leaderboard report
Dichotomous Image Segmentation DIS-VD GateNet S-Measure 0.723 #20 of 24 Archive leaderboard report
Dichotomous Image Segmentation DIS-VD GateNet max F-Measure 0.678 #20 of 24 Archive leaderboard report
Dichotomous Image Segmentation DIS-VD GateNet weighted F-measure 0.574 #20 of 24 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.

Methods

Concatenated Skip ConnectionConvolutionMax PoolingReLUSpatial Pyramid PoolingU-Net

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