Looking for change? Roll the Dice and demand Attention

Abstract

Change detection, i.e. 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 reliable deep learning framework for the task of semantic change detection in very high-resolution aerial images. Our framework consists of a new loss function, new attention modules, 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 spatial and channel convolution Attention layer (the FracTAL). The new attention layer, designed specifically for vision tasks, is memory efficient, thus suitable for use in all levels of deep convolutional networks. Based on these, we introduce two new efficient self-contained feature extraction convolution units. We validate the performance of these feature extraction building blocks on the CIFAR10 reference data and compare the results with standard ResNet modules. Further, we introduce a new encoder/decoder scheme, a network macro-topology, that is tailored for the task of change detection. Our network moves away from any notion of subtraction of feature layers for identifying change. We validate our approach by showing excellent performance and achieving state of the art score (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.

arXiv comment28 pages, under review in ISPRS P&RS, 1st revision. Figures of low quality due to compression for arxiv. Reduced abstract in arxiv due to character limitations

References in corpus (15)

• Adam: A Method for Stochastic Optimization
• Neural Machine Translation by Jointly Learning to Align and Translate
• Deep Residual Learning for Image Recognition
• U${}^2$-Net: Going Deeper with Nested U-Structure for Salient Object Detection
• ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data
• MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
• On the Properties of Neural Machine Translation: Encoder-Decoder Approaches
• DASNet: Dual attentive fully convolutional siamese networks for change detection of high resolution satellite images
• Stacked Hourglass Networks for Human Pose Estimation
• Horovod: fast and easy distributed deep learning in TensorFlow
• Recent Advances in Autoencoder-Based Representation Learning
• Identity Mappings in Deep Residual Networks
• Structured Attention Networks
• Learning to Measure Change: Fully Convolutional Siamese Metric Networks for Scene Change Detection
• SCA-CNN: Spatial and Channel-wise Attention in Convolutional Networks for Image Captioning