Vision / Model
AnomalyCLIP, trained on VisA
- Organisation
- Zhejiang University / Singapore Management University et al.
- Family
- AnomalyCLIP · version checkpoints/9_12_4_multiscale/epoch_15.pth
- Released
- 2023-10-29 (arXiv v1 submission date (https://arxiv.org/abs/2310.18961))
- Checkpoints
- checkpoints/9_12_4_multiscale/epoch_15.pth
- Weights
- Open · MIT
- Links
- PaperCode
Training data disclosure: Full list
Level determined for 12% of benchmark cells.
- TestTest data seen1
- SeenSeen, split unclear0
- TrainTraining data seen0
- Related data seen0
- NoNot seen0
- ?Unknown7
“As for MVTec AD, we fine-tune AomalyCLIP on test data on VisA.”
Built on
Everything these base models may have seen is inherited by AnomalyCLIP, trained on VisA.
-
OpenAI CLIP
· disclosure: Partial list
Evidence
“We use the publicly available CLIP model2 (VIT-L/14@336px) as our backbone. Model parameters of CLIP are all frozen.”
arxiv.org · paper · Section 4.1 · retrieved 2026-10-15 “"ViT-L/14@336px": "https://openaipublic.azureedge.net/clip/models/3035c92b350959924f9f00213499208652fc7ea050643e8b385c2dac08641f02/ViT-L-14-336px.pt"”
raw.githubusercontent.com · code · AnomalyCLIP_lib/model_load.py · retrieved 2026-10-15
Other declared training data
-
VisA · split: test · fine-tuning
Evidence
“As for MVTec AD, we fine-tune AomalyCLIP on the test data of VisA.”
arxiv.org · paper · Section 4.1 · retrieved 2026-10-15 “python train.py --dataset visa --train_data_path /remote-home/iot_zhouqihang/data/Visa”
raw.githubusercontent.com · code · train.sh · retrieved 2026-10-15
Benchmark cells
Industrial anomaly detection
Unknown MVTec AD · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Unknown MVTec AD 2 · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Unknown MVTec LOCO AD · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Test data seen VisA · in fine-tuning data: its test split was used for training
Evidence (2)
“As for MVTec AD, we fine-tune AomalyCLIP on the test data of VisA.”
“python train.py --dataset visa --train_data_path /remote-home/iot_zhouqihang/data/Visa”
Unknown Real-IAD · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Unknown MPDD · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Unknown BTAD · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Unknown KolektorSDD2 · unknown for its base model OpenAI CLIP: training data not disclosed in enough detail
Notes
Trained on the VisA test set; used only for the reported MVTec AD results. Repo folder '9_12_4_multiscale' (train.sh trains it with --dataset visa; test.sh evaluates it with --dataset mvtec).
Sources
- AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detection (ICLR 2024) · paper · retrieved 2026-10-15
- AnomalyCLIP README · readme · retrieved 2026-10-15
- train.sh · code · retrieved 2026-10-15
- model_load.py · code · retrieved 2026-10-15