Detecting Media Clones in Cultural Repositories Using a Positive Unlabeled Learning Approach

V. Sevetlidis; V. Arampatzakis; M. Karta; I. Mourthos; D. Tsiafaki; G. Pavlidis

cs.CVApr 5, 2026

Detecting Media Clones in Cultural Repositories Using a Positive Unlabeled Learning Approach

Authors: V. Sevetlidis, V. Arampatzakis, M. Karta, I. Mourthos, D. Tsiafaki, G. Pavlidis

Organizations: Archimedes, ATHENA RC, Greece

Abstract

We formulate curator-in-the-loop duplicate discovery in the AtticPOT repository as a Positive-Unlabeled (PU) learning problem. Given a single anchor per artefact, we train a lightweight per-query Clone Encoder on augmented views of the anchor and score the unlabeled repository with an interpretable threshold on the latent l_2 norm. The system proposes candidates for curator verification, uncovering cross-record duplicates that were not verified a priori. On CIFAR-10 we obtain F1=96.37 (AUROC=97.97); on AtticPOT we reach F1=90.79 (AUROC=98.99), improving F1 by +7.70 points over the best baseline (SVDD) under the same lightweight backbone. Qualitative "find-similar" panels show stable neighbourhoods across viewpoint and condition. The method avoids explicit negatives, offers a transparent operating point, and fits de-duplication, record linkage, and curator-in-the-loop workflows.

Figures & tables

Figure 1: Proposed clone-detection workflow: an anchor is augmented into clones and contrasted with an unlabeled pool; a Positive–Unlabeled–trained encoder discards non-matches and flags near-duplicates.

Figure 2: Training batch illustration on CIFAR–10. Panels A: the anchor image, B: clones of the anchor via augmentation and C: unlabeled pool mixing unlabeled positives.

Panel of Figure 3 — Figure 3: Duplicate detection performance on CIFAR-10 (left) and AtticPOT (right).

Figure 4: Qualitative retrieval on AtticPOT. For each query (leftmost image in a row), we show the nine highest-ranked images (green borders) and the single least similar (rightmost, red border).

Variant	$d$	$\lambda_{\mathrm{var}}$	$m$	WD	F1 op	AUROC	AUPRC	F1 best
L2 + learned $m$	128	0.1	learned	0	0.955	0.998	0.995	0.991
L2 + fixed $m$	64	0.0	0.500	0	0.974	0.998	0.996	0.993
L2 + fixed $m$ + WD	128	0.1	0.500	1e-04	0.971	0.999	0.998	0.994
L2 + learned $m$ + $\lambda_{\mathrm{var}}$	64	0.1	learned	0	0.954	0.994	0.986	0.987
Cosine to centroid (best-F1)	128	0.1	0.5	1e-4	–	0.987	0.985	0.962

Table 1: Ablations on CIFAR–10 proxy (mean over anchors).

Panel of Figure 5 — Figure 5: Positive vs. negative latent norm distributions for a sample anchor (left). Across anchors: distribution of $\mu$ (center) and learned $m$ (right). $\mu$ varies with the anchor.

Panel of Figure 6 — Figure 6: Calibration around the operating threshold $\tau{=}\mu{+}m$ (averaged over anchors), with $\delta\!\in\![-0.5,0.5]$ .

Explore similar work

Identity-Duplication Auditing in National-Scale Neuroimaging Repositories
Oct 7, 2026Jiheng Li, Michael E. Kim, Trent M. Schwartz +19Neuroimaging Brain MRI
HABS-HD · Department of Electrical and Computer Engineering, Vanderbilt University, Nashville, TN, USA · AI2D Center for AI and Data Science for Integrated Diagnostics, University of Pennsylvania, Philadelphia, PA, USA +2
Authentication of Copy Detection Patterns via Cross-Camera Dual-Synthetic Referencing
May 29, 2026Ivan Oleksiyuk, Roman Chaban, Slava VoloshynovskiyDomain Adaptation Digital Image Forensics
Department of Computer Science, University of Geneva, Switzerland
MINER: Multi-crop INference-time Enhancement for Rare-Object Retrieval with Frozen Dual Encoders
Sep 22, 2026Abdulmalik Alquwayfili, Faisal AlMeshal, Jumanah Almajnouni +2Vision-Language Retrieval Image Retrieval
Saudi Data and Artificial Intelligence Authority (SDAIA)

Oct 7, 2026J/K move · Enter open · S save

Identity-Duplication Auditing in National-Scale Neuroimaging Repositories

Jiheng Li, Michael E. Kim, Trent M. Schwartz, Yuhan Cui+18

HABS-HD · Department of Electrical and Computer Engineering, Vanderbilt University, Nashville, TN, USA · AI2D Center for AI and Data Science for Integrated Diagnostics, University of Pennsylvania, Philadelphia, PA, USA+2

Neuroimaging Brain MRI

Open paper AskPDF arXiv

National-scale magnetic resonance imaging (MRI) repositories increasingly integrate data from different studies and institutions. However, subject identifiers assigned to study participants are typically specific to individual datasets, so the same person may appear under different identifiers when these datasets are aggregated, a problem we define as identity duplication. Such duplication can create leakage between training and test data and inflate apparent performance in downstream biomedical studies. Existing methods do not provide an end-to-end, image-based workflow to address this problem. In this work, we present HAPPEN, a human-in-the-loop pipeline for auditing identity duplication in T1-weighted brain MRI repositories. It combines SHA-256 fingerprinting for exact-duplicate detection with embeddings learned through supervised contrastive learning to retrieve near-duplicate candidates: pairs of non-identical scans that may originate from the same person. Retrieved pairs are reviewed as candidates in a locally hosted interface rather than automatically classified as duplicates. We deployed the workflow in a national-scale aggregated repository containing 95,129 scans and assessed end-to-end recovery of 54 pairs of subject records identified as duplicates using genome-wide genotype data. Transferability was assessed by locally deploying the same workflow on 22,386 scans at an independent institution without model retraining or image transfer. Deployment in the study repository identified 1,316 exact-duplicate scan groups and 1,275 near-duplicate groups of subject records supported by human review. Of these groups, 56% and 82%, respectively, crossed dataset boundaries. All 54 genetic-reference pairs were recovered. The external team independently completed the full workflow using a locally selected operating threshold and review standard.

Abstract

Explore similar work

Identity-Duplication Auditing in National-Scale Neuroimaging Repositories

Authentication of Copy Detection Patterns via Cross-Camera Dual-Synthetic Referencing

MINER: Multi-crop INference-time Enhancement for Rare-Object Retrieval with Frozen Dual Encoders