Contamination Map Methodology

Vision / Model

AnomalyDINO

Organisation
University of Tuebingen et al. (Damm et al.)
Family
AnomalyDINO · version AnomalyDINO-S (default)
Released
2024-05-23 (arXiv v1 submission date (https://arxiv.org/abs/2405.14529))
Weights
Unknown · Apache-2.0
Links
PaperCode

Training data disclosure: Full list

Level determined for 0% of benchmark cells.

“The approach is methodologically simple and training-free and, thus, does not require any additional data for fine-tuning or meta-learning.”
arxiv.org · paper · Abstract · retrieved 2026-10-15

Built on

Everything these base models may have seen is inherited by AnomalyDINO.

Benchmark cells

Industrial anomaly detection

Unknown MVTec AD · unknown for its base model DINOv2: training data not disclosed in enough detail

Unknown MVTec AD 2 · unknown for its base model DINOv2: training data not disclosed in enough detail

Unknown MVTec LOCO AD · unknown for its base model DINOv2: training data not disclosed in enough detail

Unknown VisA · unknown for its base model DINOv2: training data not disclosed in enough detail

Unknown Real-IAD · unknown for its base model DINOv2: training data not disclosed in enough detail

Unknown MPDD · unknown for its base model DINOv2: training data not disclosed in enough detail

Unknown BTAD · unknown for its base model DINOv2: training data not disclosed in enough detail

Unknown KolektorSDD2 · unknown for its base model DINOv2: training data not disclosed in enough detail

Notes

Training-free few-shot patch nearest-neighbour detector on frozen DINOv2 (default ViT-S). Reference (few-shot) images come from the target dataset's normal training split; masking decision is made on the first reference sample.

Sources