Looking for Change? Roll the Dice and Demand Attention

Change detection, i.e., the identification per pixel of changes for some classes of interest from a set of bi-temporal co-registered images, is a fundamental task in the field of remote sensing. It remains challenging due to unrelated forms of change that appear at different times in input images. Here, we propose a deep learning framework for the task of semantic change detection in very high-resolution aerial images. Our framework consists of a new loss function, a new attention module, new feature extraction building blocks, and a new backbone architecture that is tailored for the task of semantic change detection. Specifically, we define a new form of set similarity that is based on an iterative evaluation of a variant of the Dice coefficient. We use this similarity metric to define a new loss function as well as a new, memory efficient, spatial and channel convolution Attention layer: the FracTAL. We introduce two new efficient self-contained feature extraction convolution units: the CEECNet and FracTALResNet units. Further, we propose a new encoder/decoder scheme, a network macro-topology, that is tailored for the task of change detection. The key insight in our approach is to facilitate the use of relative attention between two convolution layers in order to fuse them. We validate our approach by showing excellent performance and achieving state-of-the-art scores (F1 and Intersection over Union-hereafter IoU) on two building change detection datasets, namely, the LEVIRCD (F1: 0.918, IoU: 0.848) and the WHU (F1: 0.938, IoU: 0.882) datasets.

Automated summary

This study presents a deep learning framework for semantic change detection in high-resolution aerial images, featuring new building blocks, loss function, attention module, and backbone architecture tailored for this task. The key insight is to facilitate relative attention fusion between two convolution layers.