Multi-label Aerial Image Classification using A Bidirectional Class-wise Attention Network

Abstract

Multi-label aerial image classification is of great significance in remote sensing community, and many researches have been conducted over the past few years. However, one common limitation shared by existing methods is that the co-occurrence relationship of various classes, so called class dependency, is underexplored and leads to an inconsiderate decision. In this paper, we propose a novel end-to-end network, namely class-wise attention-based convolutional and bidirectional LSTM network (CA-Conv-BiLSTM), for this task. The proposed network consists of three indispensable components: 1) a feature extraction module, 2) a class attention learning layer, and 3) a bidirectional LSTM-based sub-network. Experimental results on UCM multi-label dataset and DFC15 multi-label dataset validate the effectiveness of our model quantitatively and qualitatively.