Convolutional autoencoder-based multimodal one-class classification
This work addresses anomaly detection in multimodal image data, such as for environmental monitoring, but it is incremental as it builds on existing autoencoder-based one-class classification techniques.
The paper tackles one-class classification for multimodal data by proposing a method using two jointly trained convolutional autoencoders to reconstruct positive data and compact latent representations, with anomaly scoring based on distance to the origin. Experimental results on a macroinvertebrate image dataset show the multimodal approach outperforms unimodal methods, and feature diversity regularizers further improve performance.
One-class classification refers to approaches of learning using data from a single class only. In this paper, we propose a deep learning one-class classification method suitable for multimodal data, which relies on two convolutional autoencoders jointly trained to reconstruct the positive input data while obtaining the data representations in the latent space as compact as possible. During inference, the distance of the latent representation of an input to the origin can be used as an anomaly score. Experimental results using a multimodal macroinvertebrate image classification dataset show that the proposed multimodal method yields better results as compared to the unimodal approach. Furthermore, study the effect of different input image sizes, and we investigate how recently proposed feature diversity regularizers affect the performance of our approach. We show that such regularizers improve performance.