Lightweight Target Detection Network

In this paper, we deeply explore the lightweight target detection network, elaborate its importance in multi-scene applications, analyze the defects of previous methods, and introduce the improvements achieved by the innovative methods. By introducing lightweight network structure, improving feature fusion mechanism and other optimization means, the lightweight transformation is based on Mask R-CNN. Experiments are conducted on the COCO dataset, and the performance of the model is evaluated using metrics such as mAP. The results show that the optimized model significantly improves the operation efficiency while maintaining high detection accuracy, providing an effective solution for target detection in resource-constrained environments.