Skip to content

About

In this repository we expand on the experiments done by Luo et. al, 2025 "Exploring Intrinsic Normal Prototypes within a Single Image for Universal Anomaly Detection". Reproduction and extension of a CVPR 2025 anomaly-detection method: MVTec AD2, LoRA fine-tuning, ONNX export and FLOPs profiling.

Topics

Resources

Stars

2 stars

Watchers

0 watching

Forks

Latest commit

 

History

12 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Reproducibility Study: INP-Former (CVPR 2025)

Reproduction and extended analysis of INP-Former: Exploring Intrinsic Normal Prototypes within a Single Image for Universal Anomaly Detection (Luo et al., CVPR 2025).

Course: MLDS2
Paper: See paper/main.tex

Repository Structure

├── INP_Former_Multi_Class.py       # Multi-class training/eval (main)
├── INP_Former_Single_Class.py      # Single-class training/eval
├── INP_Former_Few_Shot.py          # Few-shot training/eval
├── INP_Former_Super_Multi_Class.py # Super multi-class training/eval
├── INP_Former_Zero_Shot.py         # Zero-shot evaluation
├── dataset.py                      # Dataset loading
├── utils.py                        # Utilities, evaluation, losses
├── lora.py                         # [OURS] LoRA adapter implementation
├── lora_finetune.py                # [OURS] LoRA cross-dataset transfer
├── anomaly_area_analysis.py        # [OURS] Anomaly area vs performance
├── eval_aupro_strict.py            # [OURS] AU-PRO at strict FPR
├── lightshift_analysis.py          # [OURS] Lighting robustness eval
├── inference_benchmark.py          # [OURS] Throughput and latency
├── aggregate_results.py            # [OURS] Collect results into tables
├── models/               # Model architecture
├── backbones/            # Encoder weights (auto-downloaded)
├── dinov2/               # DINOv2 backbone (default)
├── optimizers/           # StableAdamW
├── scripts/              # SLURM job scripts (also runnable with bash)
├── data/                 # Local datasets (ignored by Git)
├── containers/           # Local Enroot image (ignored by Git)
├── logs/                 # Setup and SLURM logs (ignored by Git)
├── paper/                # Reproducibility report (TMLR format)
├── assets/               # Figures from the original paper
└── EXPERIMENTS.md        # Experiment tracker with all results

Files marked [OURS] are our additions; everything else is from the official INP-Former repo.

Setup

Environment

conda create -n INP python=3.8.12
conda activate INP
pip install -r requirements.txt

Tested on NVIDIA A100 (80GB) and L4 (24GB).

Backbone Weights

The DINOv2 Register ViT-Base/14 backbone is downloaded automatically on first run via torch.hub. Weights are cached in backbones/. No manual download needed.

Datasets

Run all setup and experiment commands from the repository root. Datasets and the container stay inside the checkout and are ignored by Git:

Dataset Source Target path
MVTec-AD Voxel51 mirror data/mvtec_ad/
VisA amazon-science/spot-diff data/visa/1cls/
Real-IAD realiad4ad.github.io data/Real-IAD/
MVTec-AD2 mvtec.com data/mvtec_ad2/

On a Slurm cluster, submit the setup pipeline from the repository root. If the cluster has the Pyxis/Enroot plugin, a container is built first and MVTec AD and VisA wait for it; otherwise the container step is skipped and every job uses the active Python environment. MVTec AD 2 downloads independently, and the smoke test waits for MVTec AD and VisA:

cd /path/to/Repro-INPFormer
bash scripts/submit_setup.sh

To resume after only some jobs were submitted, pass their active job IDs back to the helper, for example: CONTAINER_JOB=12345 AD2_JOB=12346 bash scripts/submit_setup.sh.

MVTec AD 2 download links can expire. Obtain the eight current category links from the official MVTec AD 2 page, then provide them as MVTECAD2_<CATEGORY>_URL environment variables. For example:

export MVTECAD2_CAN_URL='<current can URL>'
export MVTECAD2_FABRIC_URL='<current fabric URL>'
# Also set FRUIT_JELLY, RICE, SHEET_METAL, VIAL, WALLPLUGS, and WALNUTS.
sbatch --export=ALL scripts/extract_mvtecad2.sh

All eight variables are needed for a new download. Categories already present and complete under data/mvtec_ad2/ are skipped.

Each setup job stops on the first error, writes a timestamped log under logs/, and ends with SETUP SUCCESS or SETUP FAILED: <reason>. The VisA job checks the official archive SHA-256 and applies Amazon's official 1cls.csv split.

Reproducing Results

1. Main Reproduction (Multi-Class)

# MVTec-AD
python INP_Former_Multi_Class.py --dataset MVTec-AD \
    --data_path data/mvtec_ad --phase train

# VisA
python INP_Former_Multi_Class.py --dataset VisA \
    --data_path data/visa/1cls --phase train

Expected results (I-AUROC / P-AUROC / P-AUPRO):

  • MVTec-AD: 99.67 / 98.45 / 94.95
  • VisA: 98.90 / 98.97 / 95.27

2. Loss/Module Ablation (5 seeds)

Ablation on MVTec-AD with seeds {1, 2, 3, 42, 123}:

# No INP baseline (y=0, lambda=0)
python INP_Former_Multi_Class.py --dataset MVTec-AD \
    --data_path data/mvtec_ad --phase train \
    --INP_num 0 --sm_gamma 0 --seed 1

# INP + Lsm only (y=3, lambda=0)
python INP_Former_Multi_Class.py --dataset MVTec-AD \
    --data_path data/mvtec_ad --phase train \
    --sm_gamma 3 --coherence_lambda 0 --seed 1

# Full model (y=3, lambda=0.2) — default
python INP_Former_Multi_Class.py --dataset MVTec-AD \
    --data_path data/mvtec_ad --phase train --seed 1

Or via SLURM: sbatch scripts/ablation_loss_final.sh

3. LoRA Transfer Learning

python lora_finetune.py --dataset MVTec-AD \
    --data_path data/mvtec_ad \
    --pretrained_weights saved_results/.../model.pth \
    --lora_rank 4

4. Analysis Scripts

# Anomaly area vs detection performance
python anomaly_area_analysis.py --dataset MVTec-AD \
    --data_path data/mvtec_ad \
    --weights saved_results/.../model.pth

# AU-PRO at strict FPR (0.05)
python eval_aupro_strict.py --dataset MVTec-AD \
    --data_path data/mvtec_ad \
    --weights saved_results/.../model.pth

# Lighting robustness (requires MVTec-AD2)
python lightshift_analysis.py \
    --data_path data/mvtec_ad2 \
    --load_from saved_results/.../model.pth

# Inference throughput
python inference_benchmark.py --weights saved_results/.../model.pth

SLURM

All job scripts are in scripts/ and must be submitted from the repository root (or with INPFORMER_ROOT set). They request generic resources only; add your cluster's partition, account, or GPU type on the command line, e.g. sbatch -p gpu --gres=gpu:a100:1 scripts/test_all_multiclass.sh.

Each script runs its workload through scripts/env.sh: inside the Enroot image containers/inpformer_env.sqfs when Pyxis is available and the image exists, and in the current Python environment otherwise. Set INPFORMER_CONTAINER=none to force the native environment, or INPFORMER_CONTAINER_MOUNTS=/data:/data to expose datasets stored outside the checkout to the container. Most scripts also run without Slurm, e.g. bash scripts/test_all_multiclass.sh.

sbatch scripts/setup_inpformer_env.sh        # Build container (Pyxis clusters only)
sbatch scripts/download_mvtec.sh             # Download MVTec AD
sbatch scripts/download_visa.sh              # Download and prepare VisA
sbatch scripts/extract_mvtecad2.sh           # Download MVTec AD 2
sbatch scripts/smoke_test.sh                  # One-GPU setup smoke test
sbatch scripts/test_all_multiclass.sh        # Main reproduction
sbatch scripts/ablation_loss_final.sh        # Ablation with seeds
sbatch scripts/lora_finetune.sh              # LoRA experiments
sbatch scripts/epoch_efficiency.sh           # Convergence analysis

See EXPERIMENTS.md for full experiment status and results.

Our Contributions

  1. Loss/module ablation with statistical robustness — 5 seeds per config, including a novel INP+Lsm configuration not tested in the original paper
  2. LoRA transfer learning — parameter-efficient cross-dataset adaptation at ranks 2, 4, 8
  3. Epoch efficiency analysis — convergence profiling showing 99.6% of final performance at epoch 50/200
  4. Anomaly area analysis — per-bin I-AUROC and P-AUPRO revealing opposite trends: detection improves with anomaly size but localization degrades
  5. Lighting robustness — evaluation under MVTec-AD2 lighting shifts with per-category breakdown
  6. AU-PRO strict — evaluation at FPR ≤ 0.05 threshold

Citation

Original paper:

@InProceedings{luo2025INP-Former,
    author    = {Luo, Wei and Cao, Yunkang and Yao, Haiming and Zhang, Xiaotian and Lou, Jianan and Cheng, Yuqi and Shen, Weiming and Yu, Wenyong},
    title     = {Exploring Intrinsic Normal Prototypes within a Single Image for Universal Anomaly Detection},
    booktitle = {Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR)},
    month     = {June},
    year      = {2025},
    pages     = {9974-9983}
}

Acknowledgments

Based on the official INP-Former implementation by Wei Luo.

About

In this repository we expand on the experiments done by Luo et. al, 2025 "Exploring Intrinsic Normal Prototypes within a Single Image for Universal Anomaly Detection". Reproduction and extension of a CVPR 2025 anomaly-detection method: MVTec AD2, LoRA fine-tuning, ONNX export and FLOPs profiling.

Topics

Resources

Stars

2 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages