Papers › UniNet: A Contrastive Learning-guided Unified Framework with Feature Selection for...

UniNet: A Contrastive Learning-guided Unified Framework with Feature Selection for Anomaly Detection

28 Feb 2025CVPR 2025 1archive 2025-07-28

Shun Wei, Jielin Jiang, Xiaolong Xu

Anomaly detection (AD) is a crucial visual task aimed at recognizing abnormal pattern within samples. However, most existing AD methods suffer from limited generalizability, as they are primarily designed for domain-specific applications, such as industrial scenarios, and often perform poorly when applied to other domains. This challenge largely stems from the inherent discrepancies in features across domains. To bridge this domain gap, we introduce UniNet, a generic unified framework that incorporates effective feature selection and contrastive learning-guided anomaly discrimination. UniNet comprises student-teacher models and a bottleneck, featuring several vital innovations: First, we propose domain-related feature selection, where the student is guided to select and focus on representative features from the teacher with domain-relevant priors, while restoring them effectively. Second, a similarity contrastive loss function is developed to strengthen the correlations among homogeneous features. Meanwhile, a margin loss function is proposed to enforce the separation between the similarities of abnormality and normality, effectively improving the model's ability to discriminate anomalies. Third, we propose a weighted decision mechanism for dynamically evaluating the anomaly score to achieve robust AD. Large-scale experiments on 12 datasets from various domains show that UniNet surpasses existing methods.

PaperPDFConference PDFCode

Code

pangdatangtt/UniNet mentioned in paperpytorch 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

Anomaly DetectionImage ClassificationMedical Image SegmentationMulti-class Anomaly DetectionRetinal OCT Disease Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Anomaly Detection BTAD UniNet Detection AUROC 97.73 #1 of 15 Archive leaderboard report
Anomaly Detection BTAD UniNet Segmentation AUPRO 80.01 #1 of 15 Archive leaderboard report
Anomaly Detection BTAD UniNet Segmentation AUROC 97.70 #1 of 15 Archive leaderboard report
Anomaly Detection MVTec 3D-AD (RGB) UniNet Detection AUROC 95.76 #1 of 2 Archive leaderboard report
Anomaly Detection MVTec 3D-AD (RGB) UniNet Segmentation AUPRO 95.55 #1 of 2 Archive leaderboard report
Anomaly Detection MVTec AD UniNet Detection AUROC 99.90 #2 of 148 Archive leaderboard report
Anomaly Detection MVTec AD UniNet Segmentation AUPRO 96.00 #2 of 148 Archive leaderboard report
Anomaly Detection MVTec AD UniNet Segmentation AUROC 98.81 #2 of 148 Archive leaderboard report
Anomaly Detection UCSD Ped2 UniNet AUC 97.9 #7 of 14 Archive leaderboard report
Anomaly Detection VisA UniNet Detection AUROC 99.8 #1 of 50 Archive leaderboard report
Anomaly Detection VisA UniNet Segmentation AUPRO 93.9 #1 of 50 Archive leaderboard report
Anomaly Detection VisA UniNet Segmentation AUPRO (until 30% FPR) 93.9 #1 of 50 Archive leaderboard report
Anomaly Detection VisA UniNet Segmentation AUROC 98.8 #1 of 50 Archive leaderboard report
Anomaly Detection VisA UniNet(model-unified multi-class) Detection AUROC 99.15 #3 of 50 Archive leaderboard report
Anomaly Detection VisA UniNet(model-unified multi-class) F1-Score 98.29 #3 of 50 Archive leaderboard report
Image Classification ISIC2018 UniNet Accuracy 100.0 #1 of 4 Archive leaderboard report
Image Classification ISIC2018 UniNet F1 100.0 #1 of 4 Archive leaderboard report
Medical Image Segmentation CVC-ClinicDB UniNet mIoU 0.895 #17 of 48 Archive leaderboard report
Medical Image Segmentation CVC-ClinicDB UniNet mean Dice 0.942 #17 of 48 Archive leaderboard report
Medical Image Segmentation CVC-ColonDB UniNet mIoU 0.856 #5 of 25 Archive leaderboard report
Medical Image Segmentation CVC-ColonDB UniNet mean Dice 0.919 #5 of 25 Archive leaderboard report
Medical Image Segmentation Kvasir-SEG UniNet mIoU 0.857 #30 of 58 Archive leaderboard report
Medical Image Segmentation Kvasir-SEG UniNet mean Dice 0.915 #30 of 58 Archive leaderboard report
Retinal OCT Disease Classification OCT2017 UniNet Acc 100.0 #1 of 16 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

Feature SelectionFocus

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