Learning deep structured network for weakly supervised change detection

TL;DR

This paper introduces a weakly supervised deep learning method for change detection that uses only image-level labels, combining CNNs and CRFs to accurately localize changes in image pairs.

Contribution

It proposes a novel deep structured network with DAG topology and an EM framework for weakly supervised change detection, reducing the need for pixel-level annotations.

Findings

Superior detection and localization performance on benchmark datasets

Effective integration of CNN and CRF for pixel-level change estimation

Iterative EM framework improves model accuracy

Abstract

Conventional change detection methods require a large number of images to learn background models or depend on tedious pixel-level labeling by humans. In this paper, we present a weakly supervised approach that needs only image-level labels to simultaneously detect and localize changes in a pair of images. To this end, we employ a deep neural network with DAG topology to learn patterns of change from image-level labeled training data. On top of the initial CNN activations, we define a CRF model to incorporate the local differences and context with the dense connections between individual pixels. We apply a constrained mean-field algorithm to estimate the pixel-level labels, and use the estimated labels to update the parameters of the CNN in an iterative EM framework. This enables imposing global constraints on the observed foreground probability mass function. Our evaluations on four benchmark datasets demonstrate superior detection and localization performance.