vaas 0.1.7

VAAS (Vision-Attention Anomaly Scoring) is a dual-module vision framework for image anomaly detection and localisation.

VAAS — Vision-Attention Anomaly Scoring

VAAS (Vision-Attention Anomaly Scoring) is an inference-focused Python library for image anomaly detection and localisation.

It combines global attention-based reasoning with patch-level consistency analysis to produce a continuous anomaly score alongside dense anomaly maps. The output is designed to be interpretable, spatially grounded, and suitable for research and prototyping workflows.

This package provides the runtime inference pipeline only. Pretrained model weights and research assets are hosted separately.

---

Installation

Install the VAAS inference library from PyPI:

pip install vaas

Optional dependency: PyTorch

VAAS uses lazy loading for PyTorch.

• Importing vaas does not require PyTorch
• PyTorch is required only when running inference

If PyTorch is not installed, VAAS will raise a clear runtime error when inference is invoked.

To install PyTorch (CPU, CUDA, or ROCm), follow the official guide: https://pytorch.org/get-started/locally/

---

Quick Start

from vaas.inference.pipeline import VAASPipeline
from PIL import Image

pipeline = VAASPipeline.from_pretrained(
    "OBA-Research/vaas-v1-df2023",
    device="cpu",
    alpha=0.5,
)

image = Image.open("image.jpg").convert("RGB")
result = pipeline(image)

print(result["S_H"])

Output format

{
    "S_F": float,
    "S_P": float,
    "S_H": float,
    "anomaly_map": numpy.ndarray  # shape (224, 224)
}

---

Visualisation

VAAS can optionally generate qualitative visual explanations combining:

• Patch-level anomaly heatmaps
• Global attention overlays
• A hybrid anomaly score gauge

pipeline.visualize(
    image="image.jpg",
    save_path="vaas_visualization.png",
    mode="all",
    threshold=0.5,
)

---

Model Variants

This release supports pretrained models hosted on Hugging Face.

Example:

• VAAS v1 — trained on 10% of DF2023

Additional variants and future releases may extend training scale and evaluation coverage.

---

Intended Use

VAAS is intended for:

• Image anomaly detection
• Visual integrity assessment
• Explainable inspection of irregular regions
• Research on attention-based anomaly scoring
• Prototyping anomaly-aware vision systems

The library supports CPU-only inference and GPU acceleration when available.

---

Limitations

• Trained on a subset of a single dataset
• Does not classify anomaly types
• Performance may degrade on out-of-distribution imagery

VAAS should not be used as a standalone decision-making system in high-stakes applications.

---

License

MIT License

---

Maintainers

OBA-Research

• https://github.com/OBA-Research
• https://huggingface.co/OBA-Research