Cross-Modal Feature Representation Learning and Label Graph Mining in a Residual Multi-Attentional CNN-LSTM Network for Multi-Label Aerial Scene Classification

The results of aerial scene classification can provide valuable information for urban planning and land monitoring. In this specific field, there are always a number object-level semantic classes in big remote-sensing pictures. Complex label-space makes it hard to detect all the targets perceive corresponding semantics typical scene, thereby weakening sensing ability. Even worse, preparation labeled dataset training deep networks is more difficult due multiple labels. order mine visual features make good use label dependency, we propose novel framework article, namely Cross-Modal Representation Learning Label Graph Mining-based Residual Multi-Attentional CNN-LSTM (CM-GM framework). framework, residual multi-attentional convolutional neural network developed extract image features. Moreover, labels embedded by language model then form graph which be further mapped advanced (GCN). With these cross-modal feature representations (image, text), will enhanced aligned GCN-based embeddings. After that, signals fed into bi-LSTM subnetwork according built graph. CM-GM able map both graph-based correlated space appropriately, using dependency efficiently, thus improving LSTM predictor’s Experimental show that proposed achieve higher accuracy on many multi-label benchmark datasets remote field.