Aladine Chetouani

Northwestern University · United States

Unclaimed · AI-structured profileORCID OpenAlex

Northwestern computer-vision researcher spanning image quality, saliency, media forensics, medical image segmentation

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-17) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.

Research profile

Output—

Not yet scored. The snapshot contains yearly publication counts but lacks the per-work venue and type metadata this axis requires; it would measure the volume and consistency of research output. No estimate is offered here.

Impact—

Not yet scored. H-index (22) and i10-index (53) appear in the snapshot, but the paper-level citation aggregates required for this axis are missing; the dimension would measure citation influence and uptake of the work. No score is inferred.

Focus100/100

Scored 100/100. The topic profile is tightly concentrated in computer vision: image and video quality assessment leads with 125 works, followed by image retrieval and classification (69), digital media forensic detection (52), 3D shape modeling and analysis (38), and visual attention and saliency detection (38). The two 2026 catalog papers extend this core into medical image segmentation and AI governance without breaking the visual-computing focus.

Collaboration55/100

Scored 55/100 from 15 distinct coauthors in the snapshot, with 12 recurring collaborators listed, including Nicolas Drapier, Florian Mauberger, Bin Wang, Gorkem Durak, and Elif Keles — a cluster consistent with the recent medical-imaging line. The two 2026 catalog papers place him in middle and last author positions rather than first or corresponding, suggesting team-based rather than solo output.

Activity100/100

Scored 100/100. Annual work counts rise from single digits in the late 2000s (4 in 2008, 11 in 2010) to a sustained recent peak: 19 (2021), 26 (2022), 24 (2023), 33 (2024), and 24 (2025), with 13 already recorded for 2026 and 70 works over the last three years, indicating an intense and ongoing publication pace.

Reproducibility20/100

Scored 20/100 from a single scored paper (reproMean 20 across reproPapers 1). This matches Evolve's reviews: the military-AI ethics framework is rated low reproducibility as it is empirically unvalidated, while GazeRefine's training-free gaze-prompt pipeline is rated medium, reporting 88.10% Dice on Kvasir-SEG. The evidence base is too thin to generalize.

Research summary

Aladine Chetouani is a computer-vision researcher at Northwestern University whose record, resolved from OpenAlex under ORCID 0000-0002-2066-4707, centers on image and video quality assessment (125 topic-assigned works), image retrieval and classification (69), digital media forensic detection (52), 3D shape modeling and analysis (38), and visual attention and saliency detection (38). The snapshot reports an h-index of 22, an i10-index of 53, and 70 works in the last three years, with annual counts climbing from 12 (2020) to 33 (2024) and 24 (2025). Two 2026 arXiv papers reviewed by Evolve show the range of his current work: as a middle author he contributed to 'The Ethics of Artificial Intelligence in Military Operations,' which frames military AI ethics as epistemic and proposes tiered governance, secret audits, a NATO standard, and a criticality score; and as last author on 'GazeRefine,' a training-free method that uses expert gaze as a test-time prompt for DINOv3-based medical image segmentation, reaching 88.10% Dice on Kvasir-SEG — near nnU-Net — though trailing on prostate MRI. Radar dimensions score activity (100) and focus (100) at the maximum, collaboration at 55, and reproducibility at 20, based on one scored paper; output and impact axes remain unscored for lack of citation-level data. Evolve finds both papers promising but modestly validated.

Collaboration pattern

The snapshot lists 15 distinct coauthors and 12 frequent collaborators, including Nicolas Drapier, Florian Mauberger, Aurelien Chateigner, Bin Wang, Gorkem Durak, Elif Keles, and Marouane Tliba — several clustered around the medical image analysis line. In the two 2026 catalog papers he appears as a middle author and a last author rather than first or corresponding, consistent with senior co-authorship on team-driven projects; the collaboration dimension scores 55/100.

Research keywords

Image and video quality assessmentVisual attention and saliency detectionImage retrieval and classificationDigital media forensic detection3D shape modeling and analysisMedical image segmentationZero-shot segmentationTest-time promptingFoundation models (DINOv3)Gaze guidanceMilitary AI ethicsAI governance

Highlights

Training-free gaze-prompted segmentation rivals heavy baselines
As last author on GazeRefine (2026), he helped show that expert eye-gaze can serve as a test-time prompt for a training-free DINOv3 segmentation pipeline, reaching 88.10% Dice on Kvasir-SEG — close to nnU-Net — though performance trails on prostate MRI; Evolve calls the method promising but modestly validated (reproducibility: medium).
A tiered governance framework for military AI
As a middle author on 'The Ethics of Artificial Intelligence in Military Operations' (2026), he co-authored an argument that military AI ethics is epistemic, proposing tiered governance, secret audits, a NATO standard, and a criticality score; Evolve finds the framework compelling but empirically unvalidated (reproducibility: low).
Sustained output growth across two decades
Counts by year climb from 4 works (2008) to a peak of 33 (2024), with 70 works in the last three years and an h-index of 22 / i10-index of 53, marking a steady escalation in publication activity — though per-paper citation data for impact scoring is not yet available.

Timeline

2026Middle author on the arXiv preprint 'The Ethics of Artificial Intelligence in Military Operations', proposing tiered governance, secret audits, a NATO standard, and a criticality score for military AI.
2026Last author on the arXiv preprint 'GazeRefine', using expert gaze as a test-time prompt for training-free DINOv3 medical image segmentation (88.10% Dice on Kvasir-SEG).
2026Both 2026 preprints enter the Evolve review catalog, spanning AI governance and medical image analysis, with reproducibility rated low and medium respectively.

Papers in the Evolve catalog

Frequent co-authors

Nicolas Drapier×1 Florian Mauberger×1 Aurelien Chateigner×1 Mohammed Oussama Benyahia×1 Marouane Tliba×1 Mohamed Amine Kerkouri×1 Taifour Yousra×1 Bin Wang×1 Max Bengtsson×1 Gorkem Durak×1 Elif Keles×1 Zuheng Ming×1

About this profile

This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.

Bibliometrics are computed from OpenAlex; identity is name-resolved from public bibliographic data under ORCID 0000-0002-2066-4707.
The output and impact radar axes are unscored because citation-level data is unavailable; h-index and i10-index are quoted as-is from the snapshot.
The reproducibility score (20) rests on a single scored paper and should not be generalized.
Only two papers (both 2026 arXiv preprints) have been reviewed in the Evolve catalog; the 2026 publication count is a partial year.