Latent Noise Segmentation: How Neural Noise Leads to the Emergence of Segmentation and Grouping

TL;DR

This paper demonstrates that neural noise can be harnessed in deep neural networks to achieve unsupervised image segmentation and perceptual grouping, aligning with human perception phenomena and improving performance on specialized datasets.

Contribution

It introduces a novel approach where neural noise enables unsupervised segmentation, providing a new explanation for perceptual grouping and a method that outperforms existing models.

Findings

Neural noise can be used to separate objects in images.

Adding noise enables segmentation without training on labels.

Model achieves 24.9% better performance on perceptual grouping datasets.

Abstract

Humans are able to segment images effortlessly without supervision using perceptual grouping. Here, we propose a counter-intuitive computational approach to solving unsupervised perceptual grouping and segmentation: that they arise because of neural noise, rather than in spite of it. We (1) mathematically demonstrate that under realistic assumptions, neural noise can be used to separate objects from each other; (2) that adding noise in a DNN enables the network to segment images even though it was never trained on any segmentation labels; and (3) that segmenting objects using noise results in segmentation performance that aligns with the perceptual grouping phenomena observed in humans, and is sample-efficient. We introduce the Good Gestalt (GG) datasets -- six datasets designed to specifically test perceptual grouping, and show that our DNN models reproduce many important phenomena in human perception, such as illusory contours, closure, continuity, proximity, and occlusion. Finally, we (4) show that our model improves performance on our GG datasets compared to other tested unsupervised models by $24.9\%$. Together, our results suggest a novel unsupervised segmentation method requiring few assumptions, a new explanation for the formation of perceptual grouping, and a novel potential benefit of neural noise.