提出轻量动态卷积FMDConv，兼顾精度与速度，适合边缘设备部署。
FMDConv: Fast Multi-Attention Dynamic Convolution via Speed-Accuracy Trade-off
- 融合三重注意力机制，按需动态调整卷积核，降低计算开销。
- 在ResNet-18上减少49.8%计算量，ResNet-50上减少42.2%，精度不降。
- 设计新评估指标，系统衡量速度与精度的平衡，适合边缘场景应用。
空间卷积是构建视觉识别深度卷积神经网络的基础。尽管动态卷积通过自适应组合静态卷积核提升了模型精度，但其带来显著计算开销，限制了在联邦边缘计算等资源受限环境中的部署。为此，我们提出快速多注意力动态卷积（FMDConv），融合输入注意力、温度退化卷积核注意力和输出注意力，优化速度-精度权衡。FMDConv通过选择性增强特征提取，在更低复杂度下实现更优平衡。此外，我们引入两个新量化指标——逆效率得分与速率修正得分，系统评估该权衡。在CIFAR-10、CIFAR-100和ImageNet上的大量实验表明，相比先前的多注意力动态卷积方法，FMDConv在ResNet-18上计算成本降低高达49.8%，在ResNet-50上降低42.2%，同时保持竞争力的精度。这些优势使其极适用于真实世界中资源受限的应用场景。
原文摘要 · Abstract (English)
Spatial convolution is fundamental in constructing deep Convolutional Neural Networks (CNNs) for visual recognition. While dynamic convolution enhances model accuracy by adaptively combining static kernels, it incurs significant computational overhead, limiting its deployment in resource-constrained environments such as federated edge computing. To address this, we propose Fast Multi-Attention Dynamic Convolution (FMDConv), which integrates input attention, temperature-degraded kernel attention, and output attention to optimize the speed-accuracy trade-off. FMDConv achieves a better balance between accuracy and efficiency by selectively enhancing feature extraction with lower complexity. Furthermore, we introduce two novel quantitative metrics, the Inverse Efficiency Score and Rate-Correct Score, to systematically evaluate this trade-off. Extensive experiments on CIFAR-10, CIFAR-100, and ImageNet demonstrate that FMDConv reduces the computational cost by up to 49.8\% on ResNet-18 and 42.2\% on ResNet-50 compared to prior multi-attention dynamic convolution methods while maintaining competitive accuracy. These advantages make FMDConv highly suitable for real-world, resource-constrained applications.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。