Redefining Normal: A Novel Object-Level Approach for Multi-Object Novelty Detection

TL;DR

This paper introduces an object-level normality definition and novel knowledge distillation techniques to improve multi-object novelty detection, outperforming existing methods in complex scenarios.

Contribution

It proposes redefining normal at the object level and develops two new distillation methods, DeFeND and masked knowledge distillation, for enhanced multi-object novelty detection.

Findings

Outperforms existing methods in multi-object novelty detection

Effective in both single-object and multi-object scenarios

Improves generalization from incomplete data

Abstract

In the realm of novelty detection, accurately identifying outliers in data without specific class information poses a significant challenge. While current methods excel in single-object scenarios, they struggle with multi-object situations due to their focus on individual objects. Our paper suggests a novel approach: redefining `normal' at the object level in training datasets. Rather than the usual image-level view, we consider the most dominant object in a dataset as the norm, offering a perspective that is more effective for real-world scenarios. Adapting to our object-level definition of `normal', we modify knowledge distillation frameworks, where a student network learns from a pre-trained teacher network. Our first contribution, DeFeND(Dense Feature Fine-tuning on Normal Data), integrates dense feature fine-tuning into the distillation process, allowing the teacher network to focus on object-level features with a self-supervised loss. The second is masked knowledge distillation, where the student network works with partially hidden inputs, honing its ability to deduce and generalize from incomplete data. This approach not only fares well in single-object novelty detection but also considerably surpasses existing methods in multi-object contexts. The implementation is available at: https://github.com/SMSD75/Redefining_Normal_ACCV24/tree/main