Fake It Till You Make It: Towards Accurate Near-Distribution Novelty Detection
This addresses a critical bottleneck in novelty detection for applications like medical imaging and quality control, though it is an incremental improvement over existing generative approaches.
The paper tackles the problem of near-distribution novelty detection in images, where subtle differences cause existing methods to drop by up to 20% in performance, and proposes using a score-based generative model to synthesize anomalous data, resulting in improved performance that reduces the gap between near-distribution and standard settings.
We aim for image-based novelty detection. Despite considerable progress, existing models either fail or face a dramatic drop under the so-called "near-distribution" setting, where the differences between normal and anomalous samples are subtle. We first demonstrate existing methods experience up to 20% decrease in performance in the near-distribution setting. Next, we propose to exploit a score-based generative model to produce synthetic near-distribution anomalous data. Our model is then fine-tuned to distinguish such data from the normal samples. We provide a quantitative as well as qualitative evaluation of this strategy, and compare the results with a variety of GAN-based models. Effectiveness of our method for both the near-distribution and standard novelty detection is assessed through extensive experiments on datasets in diverse applications such as medical images, object classification, and quality control. This reveals that our method considerably improves over existing models, and consistently decreases the gap between the near-distribution and standard novelty detection performance. The code repository is available at https://github.com/rohban-lab/FITYMI.