Contamination Map Methodology

Vision / Model

Segment Any Anomaly (SAA / SAA+)

Organisation
Huazhong University of Science and Technology et al.
Family
SAA
Released
2023-05-18 (arXiv v1 submission date (https://arxiv.org/abs/2305.10724))
Weights
Unknown
Links
PaperCode

Training data disclosure: Full list

Level determined for 0% of benchmark cells.

“Notably, as all methods require no training process, their performance is stable”
arxiv.org · paper · Section 4.2 · retrieved 2026-10-15
“This project aims to segment any anomaly without any training.”
raw.githubusercontent.com · readme · Introduction · retrieved 2026-10-15

Built on

Everything these base models may have seen is inherited by Segment Any Anomaly (SAA / SAA+).

Other declared training data

Benchmark cells

Industrial anomaly detection

Unknown MVTec AD · unknown for its base model Segment Anything Model (SAM): training data not disclosed in enough detail

Unknown MVTec AD 2 · training data not disclosed in enough detail

Unknown MVTec LOCO AD · unknown for its base model Segment Anything Model (SAM): training data not disclosed in enough detail

Unknown VisA · unknown for its base model Segment Anything Model (SAM): training data not disclosed in enough detail

Unknown Real-IAD · training data not disclosed in enough detail

Unknown MPDD · unknown for its base model Segment Anything Model (SAM): training data not disclosed in enough detail

Unknown BTAD · unknown for its base model Segment Anything Model (SAM): training data not disclosed in enough detail

Unknown KolektorSDD2 · unknown for its base model Segment Anything Model (SAM): training data not disclosed in enough detail

Notes

Training-free; prompts derived from domain expert knowledge (per-category text and object-property prompts). Evaluated zero-shot on VisA, MVTec-AD, KSDD2 and MTD.

Sources