Papers › Explicit Visual Prompting for Universal Foreground Segmentations

Explicit Visual Prompting for Universal Foreground Segmentations

29 May 2023arXiv:2305.18476archive 2025-07-28

Weihuang Liu, Xi Shen, Chi-Man Pun, Xiaodong Cun

Foreground segmentation is a fundamental problem in computer vision, which includes salient object detection, forgery detection, defocus blur detection, shadow detection, and camouflage object detection. Previous works have typically relied on domain-specific solutions to address accuracy and robustness issues in those applications. In this paper, we present a unified framework for a number of foreground segmentation tasks without any task-specific designs. We take inspiration from the widely-used pre-training and then prompt tuning protocols in NLP and propose a new visual prompting model, named Explicit Visual Prompting (EVP). Different from the previous visual prompting which is typically a dataset-level implicit embedding, our key insight is to enforce the tunable parameters focusing on the explicit visual content from each individual image, i.e., the features from frozen patch embeddings and high-frequency components. Our method freezes a pre-trained model and then learns task-specific knowledge using a few extra parameters. Despite introducing only a small number of tunable parameters, EVP achieves superior performance than full fine-tuning and other parameter-efficient fine-tuning methods. Experiments in fourteen datasets across five tasks show the proposed method outperforms other task-specific methods while being considerably simple. The proposed method demonstrates the scalability in different architectures, pre-trained weights, and tasks. The code is available at: https://github.com/NiFangBaAGe/Explicit-Visual-Prompt.

PaperPDFCode

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

Code

nifangbaage/explicit-visual-prompt officialmentioned in papermentioned on GitHubpytorch report
nifangbaage/explict-visual-prompt 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

Camouflaged Object SegmentationDefocus Blur DetectionForeground SegmentationImage Manipulation DetectionSalient Object DetectionShadow DetectionVisual Promptingparameter-efficient fine-tuning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Camouflaged Object Segmentation CAMO EVPv2 MAE 0.058 #5 of 14 Archive leaderboard report
Camouflaged Object Segmentation CAMO EVPv2 S-Measure 0.848 #5 of 14 Archive leaderboard report
Camouflaged Object Segmentation CAMO EVPv2 Weighted F-Measure 0.786 #5 of 14 Archive leaderboard report
Camouflaged Object Segmentation COD EVPv2 MAE 0.029 #6 of 12 Archive leaderboard report
Camouflaged Object Segmentation COD EVPv2 S-Measure 0.843 #6 of 12 Archive leaderboard report
Camouflaged Object Segmentation COD EVPv2 Weighted F-Measure 0.746 #6 of 12 Archive leaderboard report
Salient Object Detection DUT-OMRON EVPv2 E-measure 0.895 #3 of 8 Archive leaderboard report
Salient Object Detection DUT-OMRON EVPv2 MAE 0.047 #3 of 8 Archive leaderboard report
Salient Object Detection DUT-OMRON EVPv2 S-measure 0.862 #3 of 8 Archive leaderboard report
Salient Object Detection DUT-OMRON EVPv2 max_F1 0.857 #3 of 8 Archive leaderboard report
Salient Object Detection DUTS-TE EVPv2 E-measure 0.948 #4 of 8 Archive leaderboard report
Salient Object Detection DUTS-TE EVPv2 MAE 0.027 #4 of 8 Archive leaderboard report
Salient Object Detection DUTS-TE EVPv2 Smeasure 0.915 #4 of 8 Archive leaderboard report
Salient Object Detection DUTS-TE EVPv2 max_F1 0.923 #4 of 8 Archive leaderboard report
Salient Object Detection ECSSD EVPv2 E-measure 0.957 #5 of 10 Archive leaderboard report
Salient Object Detection ECSSD EVPv2 MAE 0.028 #5 of 10 Archive leaderboard report
Salient Object Detection ECSSD EVPv2 S-measure 0.935 #5 of 10 Archive leaderboard report
Salient Object Detection ECSSD EVPv2 max_F1 0.958 #5 of 10 Archive leaderboard report
Salient Object Detection PASCAL-S EVPv2 E-measure 0.917 #3 of 10 Archive leaderboard report
Salient Object Detection PASCAL-S EVPv2 MAE 0.053 #3 of 10 Archive leaderboard report
Salient Object Detection PASCAL-S EVPv2 S-measure 0.879 #3 of 10 Archive leaderboard report
Salient Object Detection PASCAL-S EVPv2 max_F1 0.869 #3 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