Noise or Signal: The Role of Image Backgrounds in Object Recognition

TL;DR

This paper investigates how state-of-the-art object recognition models rely on image backgrounds, revealing that models often depend on backgrounds for classification and that reliance decreases with model accuracy.

Contribution

The authors develop a toolkit to disentangle foreground and background signals, providing insights into background dependence and its impact on model performance.

Findings

Models can classify using backgrounds alone.

High reliance on backgrounds correlates with lower accuracy.

More accurate models depend less on backgrounds.

Abstract

We assess the tendency of state-of-the-art object recognition models to depend on signals from image backgrounds. We create a toolkit for disentangling foreground and background signal on ImageNet images, and find that (a) models can achieve non-trivial accuracy by relying on the background alone, (b) models often misclassify images even in the presence of correctly classified foregrounds--up to 87.5% of the time with adversarially chosen backgrounds, and (c) more accurate models tend to depend on backgrounds less. Our analysis of backgrounds brings us closer to understanding which correlations machine learning models use, and how they determine models' out of distribution performance.