Heuristic Hyperparameter Choice for Image Anomaly Detection
This work addresses computational cost and performance degradation in image anomaly detection, but it is incremental as it builds on existing NPCA methods.
The paper tackles the problem of redundant features in image anomaly detection by proposing a heuristic method to choose hyperparameters for Negated Principal Component Analysis, achieving dimension reduction while maintaining performance.
Anomaly detection (AD) in images is a fundamental computer vision problem by deep learning neural network to identify images deviating significantly from normality. The deep features extracted from pretrained models have been proved to be essential for AD based on multivariate Gaussian distribution analysis. However, since models are usually pretrained on a large dataset for classification tasks such as ImageNet, they might produce lots of redundant features for AD, which increases computational cost and degrades the performance. We aim to do the dimension reduction of Negated Principal Component Analysis (NPCA) for these features. So we proposed some heuristic to choose hyperparameter of NPCA algorithm for getting as fewer components of features as possible while ensuring a good performance.