Papers › Gradient-Induced Co-Saliency Detection

Gradient-Induced Co-Saliency Detection

28 Apr 2020ECCV 2020 8arXiv:2004.13364archive 2025-07-28

Zhao Zhang, Wenda Jin, Jun Xu, Ming-Ming Cheng

Co-saliency detection (Co-SOD) aims to segment the common salient foreground in a group of relevant images. In this paper, inspired by human behavior, we propose a gradient-induced co-saliency detection (GICD) method. We first abstract a consensus representation for the grouped images in the embedding space; then, by comparing the single image with consensus representation, we utilize the feedback gradient information to induce more attention to the discriminative co-salient features. In addition, due to the lack of Co-SOD training data, we design a jigsaw training strategy, with which Co-SOD networks can be trained on general saliency datasets without extra pixel-level annotations. To evaluate the performance of Co-SOD methods on discovering the co-salient object among multiple foregrounds, we construct a challenging CoCA dataset, where each image contains at least one extraneous foreground along with the co-salient object. Experiments demonstrate that our GICD achieves state-of-the-art performance. Our codes and dataset are available at https://mmcheng.net/gicd/.

PaperPDFConference PDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

zzhanghub/gicd mentioned on GitHubpytorchSyntology: not harvested 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

Co-Salient Object DetectionSaliency Detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Co-Salient Object Detection CoCA GICD MAE 0.126 #8 of 10 Archive leaderboard report
Co-Salient Object Detection CoCA GICD Mean F-measure 0.504 #8 of 10 Archive leaderboard report
Co-Salient Object Detection CoCA GICD S-measure 0.658 #8 of 10 Archive leaderboard report
Co-Salient Object Detection CoCA GICD max E-measure 0.715 #8 of 10 Archive leaderboard report
Co-Salient Object Detection CoCA GICD max F-measure 0.513 #8 of 10 Archive leaderboard report
Co-Salient Object Detection CoCA GICD mean E-measure 0.701 #8 of 10 Archive leaderboard report
Co-Salient Object Detection CoSOD3k GICD MAE 0.079 #7 of 10 Archive leaderboard report
Co-Salient Object Detection CoSOD3k GICD S-measure 0.797 #7 of 10 Archive leaderboard report
Co-Salient Object Detection CoSOD3k GICD max E-measure 0.848 #7 of 10 Archive leaderboard report
Co-Salient Object Detection CoSOD3k GICD max F-measure 0.770 #7 of 10 Archive leaderboard report
Co-Salient Object Detection CoSOD3k GICD mean E-measure 0.845 #7 of 10 Archive leaderboard report
Co-Salient Object Detection CoSOD3k GICD mean F-measure 0.763 #7 of 10 Archive leaderboard report
Co-Salient Object Detection CoSal2015 GICD MAE 0.071 #7 of 10 Archive leaderboard report
Co-Salient Object Detection CoSal2015 GICD S-measure 0.844 #7 of 10 Archive leaderboard report
Co-Salient Object Detection CoSal2015 GICD max E-measure 0.887 #7 of 10 Archive leaderboard report
Co-Salient Object Detection CoSal2015 GICD max F-measure 0.844 #7 of 10 Archive leaderboard report
Co-Salient Object Detection CoSal2015 GICD mean E-measure 0.883 #7 of 10 Archive leaderboard report
Co-Salient Object Detection CoSal2015 GICD mean F-measure 0.835 #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.

Methods

Jigsaw

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