RemoteTrimmer: Adaptive Structural Pruning for Remote Sensing Image Classification

TL;DR

RemoteTrimmer introduces an adaptive structural pruning method tailored for remote sensing image classification, effectively reducing model complexity while maintaining high accuracy, and outperforming existing approaches.

Contribution

The paper presents a novel adaptive pruning strategy and a specialized loss function for fine-tuning, addressing the unique challenges of remote sensing image data.

Findings

Achieves minimal accuracy loss after pruning

Outperforms state-of-the-art methods on remote sensing datasets

Effective in maintaining model performance with reduced complexity

Abstract

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.