RemoteTrimmer: Adaptive Structural Pruning for Remote Sensing Image Classification
This work addresses the need for lightweight models in remote sensing image classification, offering a domain-specific solution that is incremental in improving pruning methods for this field.
The paper tackles the problem of high computational complexity in remote sensing image classification by proposing a structural pruning method that minimizes accuracy loss, achieving state-of-the-art performance with minimal accuracy loss on two datasets.
Since high resolution remote sensing image classification often requires a relatively high computation complexity, lightweight models tend to be practical and efficient. Model pruning is an effective method for model compression. However, existing methods rarely take into account the specificity of remote sensing images, resulting in significant accuracy loss after pruning. To this end, we propose an effective structural pruning approach for remote sensing image classification. Specifically, a pruning strategy that amplifies the differences in channel importance of the model is introduced. Then an adaptive mining loss function is designed for the fine-tuning process of the pruned model. Finally, we conducted experiments on two remote sensing classification datasets. The experimental results demonstrate that our method achieves minimal accuracy loss after compressing remote sensing classification models, achieving state-of-the-art (SoTA) performance.