ODDObjects: A Framework for Multiclass Unsupervised Anomaly Detection on Masked Objects

TL;DR

ODDObjects introduces a new unsupervised framework for detecting anomalies in masked objects using autoencoders, demonstrating superior performance with memory-augmented models on COCO-style datasets.

Contribution

The paper proposes a novel autoencoder-based framework for multiclass anomaly detection on masked objects, extending prior methods with memory-augmented architectures.

Findings

Memory-augmented deep convolutional autoencoders outperform other models in anomaly detection.

The framework effectively detects out-of-distribution objects across various categories.

Experimental results validate the approach on COCO-style datasets.

Abstract

This paper presents a novel framework for unsupervised anomaly detection on masked objects called ODDObjects, which stands for Out-of-Distribution Detection on Objects. ODDObjects is designed to detect anomalies of various categories using unsupervised autoencoders trained on COCO-style datasets. The method utilizes autoencoder-based image reconstruction, where high reconstruction error indicates the possibility of an anomaly. The framework extends previous work on anomaly detection with autoencoders, comparing state-of-the-art models trained on object recognition datasets. Various model architectures were compared, and experimental results show that memory-augmented deep convolutional autoencoders perform the best at detecting out-of-distribution objects.